2005-04-17 02:20:36 +04:00
|
|
|
/*
|
2007-11-19 12:57:03 +03:00
|
|
|
* arch/sh/mm/tlb-flush_64.c
|
2005-04-17 02:20:36 +04:00
|
|
|
*
|
|
|
|
* Copyright (C) 2000, 2001 Paolo Alberelli
|
|
|
|
* Copyright (C) 2003 Richard Curnow (/proc/tlb, bug fixes)
|
2009-06-24 21:49:03 +04:00
|
|
|
* Copyright (C) 2003 - 2009 Paul Mundt
|
2005-04-17 02:20:36 +04:00
|
|
|
*
|
2007-11-19 12:57:03 +03:00
|
|
|
* This file is subject to the terms and conditions of the GNU General Public
|
|
|
|
* License. See the file "COPYING" in the main directory of this archive
|
|
|
|
* for more details.
|
2005-04-17 02:20:36 +04:00
|
|
|
*/
|
|
|
|
#include <linux/signal.h>
|
|
|
|
#include <linux/rwsem.h>
|
|
|
|
#include <linux/sched.h>
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/errno.h>
|
|
|
|
#include <linux/string.h>
|
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/ptrace.h>
|
|
|
|
#include <linux/mman.h>
|
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/smp.h>
|
perf: Do the big rename: Performance Counters -> Performance Events
Bye-bye Performance Counters, welcome Performance Events!
In the past few months the perfcounters subsystem has grown out its
initial role of counting hardware events, and has become (and is
becoming) a much broader generic event enumeration, reporting, logging,
monitoring, analysis facility.
Naming its core object 'perf_counter' and naming the subsystem
'perfcounters' has become more and more of a misnomer. With pending
code like hw-breakpoints support the 'counter' name is less and
less appropriate.
All in one, we've decided to rename the subsystem to 'performance
events' and to propagate this rename through all fields, variables
and API names. (in an ABI compatible fashion)
The word 'event' is also a bit shorter than 'counter' - which makes
it slightly more convenient to write/handle as well.
Thanks goes to Stephane Eranian who first observed this misnomer and
suggested a rename.
User-space tooling and ABI compatibility is not affected - this patch
should be function-invariant. (Also, defconfigs were not touched to
keep the size down.)
This patch has been generated via the following script:
FILES=$(find * -type f | grep -vE 'oprofile|[^K]config')
sed -i \
-e 's/PERF_EVENT_/PERF_RECORD_/g' \
-e 's/PERF_COUNTER/PERF_EVENT/g' \
-e 's/perf_counter/perf_event/g' \
-e 's/nb_counters/nb_events/g' \
-e 's/swcounter/swevent/g' \
-e 's/tpcounter_event/tp_event/g' \
$FILES
for N in $(find . -name perf_counter.[ch]); do
M=$(echo $N | sed 's/perf_counter/perf_event/g')
mv $N $M
done
FILES=$(find . -name perf_event.*)
sed -i \
-e 's/COUNTER_MASK/REG_MASK/g' \
-e 's/COUNTER/EVENT/g' \
-e 's/\<event\>/event_id/g' \
-e 's/counter/event/g' \
-e 's/Counter/Event/g' \
$FILES
... to keep it as correct as possible. This script can also be
used by anyone who has pending perfcounters patches - it converts
a Linux kernel tree over to the new naming. We tried to time this
change to the point in time where the amount of pending patches
is the smallest: the end of the merge window.
Namespace clashes were fixed up in a preparatory patch - and some
stylistic fallout will be fixed up in a subsequent patch.
( NOTE: 'counters' are still the proper terminology when we deal
with hardware registers - and these sed scripts are a bit
over-eager in renaming them. I've undone some of that, but
in case there's something left where 'counter' would be
better than 'event' we can undo that on an individual basis
instead of touching an otherwise nicely automated patch. )
Suggested-by: Stephane Eranian <eranian@google.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Paul Mackerras <paulus@samba.org>
Reviewed-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <linux-arch@vger.kernel.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-09-21 14:02:48 +04:00
|
|
|
#include <linux/perf_event.h>
|
2005-04-17 02:20:36 +04:00
|
|
|
#include <linux/interrupt.h>
|
|
|
|
#include <asm/system.h>
|
|
|
|
#include <asm/io.h>
|
|
|
|
#include <asm/tlb.h>
|
|
|
|
#include <asm/uaccess.h>
|
|
|
|
#include <asm/pgalloc.h>
|
|
|
|
#include <asm/mmu_context.h>
|
|
|
|
|
|
|
|
extern void die(const char *,struct pt_regs *,long);
|
|
|
|
|
|
|
|
#define PFLAG(val,flag) (( (val) & (flag) ) ? #flag : "" )
|
|
|
|
#define PPROT(flag) PFLAG(pgprot_val(prot),flag)
|
|
|
|
|
|
|
|
static inline void print_prots(pgprot_t prot)
|
|
|
|
{
|
2009-11-20 00:11:05 +03:00
|
|
|
printk("prot is 0x%016llx\n",pgprot_val(prot));
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
printk("%s %s %s %s %s\n",PPROT(_PAGE_SHARED),PPROT(_PAGE_READ),
|
|
|
|
PPROT(_PAGE_EXECUTE),PPROT(_PAGE_WRITE),PPROT(_PAGE_USER));
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void print_vma(struct vm_area_struct *vma)
|
|
|
|
{
|
|
|
|
printk("vma start 0x%08lx\n", vma->vm_start);
|
|
|
|
printk("vma end 0x%08lx\n", vma->vm_end);
|
|
|
|
|
|
|
|
print_prots(vma->vm_page_prot);
|
|
|
|
printk("vm_flags 0x%08lx\n", vma->vm_flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void print_task(struct task_struct *tsk)
|
|
|
|
{
|
2007-10-19 10:40:41 +04:00
|
|
|
printk("Task pid %d\n", task_pid_nr(tsk));
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
static pte_t *lookup_pte(struct mm_struct *mm, unsigned long address)
|
|
|
|
{
|
|
|
|
pgd_t *dir;
|
2007-11-19 12:57:03 +03:00
|
|
|
pud_t *pud;
|
2005-04-17 02:20:36 +04:00
|
|
|
pmd_t *pmd;
|
|
|
|
pte_t *pte;
|
|
|
|
pte_t entry;
|
|
|
|
|
|
|
|
dir = pgd_offset(mm, address);
|
2007-11-19 12:57:03 +03:00
|
|
|
if (pgd_none(*dir))
|
2005-04-17 02:20:36 +04:00
|
|
|
return NULL;
|
|
|
|
|
2007-11-19 12:57:03 +03:00
|
|
|
pud = pud_offset(dir, address);
|
|
|
|
if (pud_none(*pud))
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
pmd = pmd_offset(pud, address);
|
|
|
|
if (pmd_none(*pmd))
|
2005-04-17 02:20:36 +04:00
|
|
|
return NULL;
|
|
|
|
|
|
|
|
pte = pte_offset_kernel(pmd, address);
|
|
|
|
entry = *pte;
|
2007-11-19 12:57:03 +03:00
|
|
|
if (pte_none(entry) || !pte_present(entry))
|
2005-04-17 02:20:36 +04:00
|
|
|
return NULL;
|
|
|
|
|
|
|
|
return pte;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* This routine handles page faults. It determines the address,
|
|
|
|
* and the problem, and then passes it off to one of the appropriate
|
|
|
|
* routines.
|
|
|
|
*/
|
|
|
|
asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
|
|
|
|
unsigned long textaccess, unsigned long address)
|
|
|
|
{
|
|
|
|
struct task_struct *tsk;
|
|
|
|
struct mm_struct *mm;
|
|
|
|
struct vm_area_struct * vma;
|
|
|
|
const struct exception_table_entry *fixup;
|
|
|
|
pte_t *pte;
|
2007-07-19 12:47:05 +04:00
|
|
|
int fault;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* SIM
|
|
|
|
* Note this is now called with interrupts still disabled
|
|
|
|
* This is to cope with being called for a missing IO port
|
2007-05-14 03:25:48 +04:00
|
|
|
* address with interrupts disabled. This should be fixed as
|
2005-04-17 02:20:36 +04:00
|
|
|
* soon as we have a better 'fast path' miss handler.
|
|
|
|
*
|
|
|
|
* Plus take care how you try and debug this stuff.
|
|
|
|
* For example, writing debug data to a port which you
|
|
|
|
* have just faulted on is not going to work.
|
|
|
|
*/
|
|
|
|
|
|
|
|
tsk = current;
|
|
|
|
mm = tsk->mm;
|
|
|
|
|
|
|
|
/* Not an IO address, so reenable interrupts */
|
|
|
|
local_irq_enable();
|
|
|
|
|
perf: Do the big rename: Performance Counters -> Performance Events
Bye-bye Performance Counters, welcome Performance Events!
In the past few months the perfcounters subsystem has grown out its
initial role of counting hardware events, and has become (and is
becoming) a much broader generic event enumeration, reporting, logging,
monitoring, analysis facility.
Naming its core object 'perf_counter' and naming the subsystem
'perfcounters' has become more and more of a misnomer. With pending
code like hw-breakpoints support the 'counter' name is less and
less appropriate.
All in one, we've decided to rename the subsystem to 'performance
events' and to propagate this rename through all fields, variables
and API names. (in an ABI compatible fashion)
The word 'event' is also a bit shorter than 'counter' - which makes
it slightly more convenient to write/handle as well.
Thanks goes to Stephane Eranian who first observed this misnomer and
suggested a rename.
User-space tooling and ABI compatibility is not affected - this patch
should be function-invariant. (Also, defconfigs were not touched to
keep the size down.)
This patch has been generated via the following script:
FILES=$(find * -type f | grep -vE 'oprofile|[^K]config')
sed -i \
-e 's/PERF_EVENT_/PERF_RECORD_/g' \
-e 's/PERF_COUNTER/PERF_EVENT/g' \
-e 's/perf_counter/perf_event/g' \
-e 's/nb_counters/nb_events/g' \
-e 's/swcounter/swevent/g' \
-e 's/tpcounter_event/tp_event/g' \
$FILES
for N in $(find . -name perf_counter.[ch]); do
M=$(echo $N | sed 's/perf_counter/perf_event/g')
mv $N $M
done
FILES=$(find . -name perf_event.*)
sed -i \
-e 's/COUNTER_MASK/REG_MASK/g' \
-e 's/COUNTER/EVENT/g' \
-e 's/\<event\>/event_id/g' \
-e 's/counter/event/g' \
-e 's/Counter/Event/g' \
$FILES
... to keep it as correct as possible. This script can also be
used by anyone who has pending perfcounters patches - it converts
a Linux kernel tree over to the new naming. We tried to time this
change to the point in time where the amount of pending patches
is the smallest: the end of the merge window.
Namespace clashes were fixed up in a preparatory patch - and some
stylistic fallout will be fixed up in a subsequent patch.
( NOTE: 'counters' are still the proper terminology when we deal
with hardware registers - and these sed scripts are a bit
over-eager in renaming them. I've undone some of that, but
in case there's something left where 'counter' would be
better than 'event' we can undo that on an individual basis
instead of touching an otherwise nicely automated patch. )
Suggested-by: Stephane Eranian <eranian@google.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Paul Mackerras <paulus@samba.org>
Reviewed-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <linux-arch@vger.kernel.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-09-21 14:02:48 +04:00
|
|
|
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, 0, regs, address);
|
2009-06-24 21:49:03 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/*
|
|
|
|
* If we're in an interrupt or have no user
|
|
|
|
* context, we must not take the fault..
|
|
|
|
*/
|
2006-12-07 07:32:18 +03:00
|
|
|
if (in_atomic() || !mm)
|
2005-04-17 02:20:36 +04:00
|
|
|
goto no_context;
|
|
|
|
|
|
|
|
/* TLB misses upon some cache flushes get done under cli() */
|
|
|
|
down_read(&mm->mmap_sem);
|
|
|
|
|
|
|
|
vma = find_vma(mm, address);
|
|
|
|
|
|
|
|
if (!vma) {
|
|
|
|
#ifdef DEBUG_FAULT
|
|
|
|
print_task(tsk);
|
|
|
|
printk("%s:%d fault, address is 0x%08x PC %016Lx textaccess %d writeaccess %d\n",
|
2008-03-05 02:23:47 +03:00
|
|
|
__func__, __LINE__,
|
2005-04-17 02:20:36 +04:00
|
|
|
address,regs->pc,textaccess,writeaccess);
|
|
|
|
show_regs(regs);
|
|
|
|
#endif
|
|
|
|
goto bad_area;
|
|
|
|
}
|
|
|
|
if (vma->vm_start <= address) {
|
|
|
|
goto good_area;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!(vma->vm_flags & VM_GROWSDOWN)) {
|
|
|
|
#ifdef DEBUG_FAULT
|
|
|
|
print_task(tsk);
|
|
|
|
printk("%s:%d fault, address is 0x%08x PC %016Lx textaccess %d writeaccess %d\n",
|
2008-03-05 02:23:47 +03:00
|
|
|
__func__, __LINE__,
|
2005-04-17 02:20:36 +04:00
|
|
|
address,regs->pc,textaccess,writeaccess);
|
|
|
|
show_regs(regs);
|
|
|
|
|
|
|
|
print_vma(vma);
|
|
|
|
#endif
|
|
|
|
goto bad_area;
|
|
|
|
}
|
|
|
|
if (expand_stack(vma, address)) {
|
|
|
|
#ifdef DEBUG_FAULT
|
|
|
|
print_task(tsk);
|
|
|
|
printk("%s:%d fault, address is 0x%08x PC %016Lx textaccess %d writeaccess %d\n",
|
2008-03-05 02:23:47 +03:00
|
|
|
__func__, __LINE__,
|
2005-04-17 02:20:36 +04:00
|
|
|
address,regs->pc,textaccess,writeaccess);
|
|
|
|
show_regs(regs);
|
|
|
|
#endif
|
|
|
|
goto bad_area;
|
|
|
|
}
|
|
|
|
/*
|
|
|
|
* Ok, we have a good vm_area for this memory access, so
|
|
|
|
* we can handle it..
|
|
|
|
*/
|
|
|
|
good_area:
|
|
|
|
if (textaccess) {
|
|
|
|
if (!(vma->vm_flags & VM_EXEC))
|
|
|
|
goto bad_area;
|
|
|
|
} else {
|
|
|
|
if (writeaccess) {
|
|
|
|
if (!(vma->vm_flags & VM_WRITE))
|
|
|
|
goto bad_area;
|
|
|
|
} else {
|
|
|
|
if (!(vma->vm_flags & VM_READ))
|
|
|
|
goto bad_area;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If for any reason at all we couldn't handle the fault,
|
|
|
|
* make sure we exit gracefully rather than endlessly redo
|
|
|
|
* the fault.
|
|
|
|
*/
|
2009-04-10 20:01:23 +04:00
|
|
|
fault = handle_mm_fault(mm, vma, address, writeaccess ? FAULT_FLAG_WRITE : 0);
|
2007-07-19 12:47:05 +04:00
|
|
|
if (unlikely(fault & VM_FAULT_ERROR)) {
|
|
|
|
if (fault & VM_FAULT_OOM)
|
|
|
|
goto out_of_memory;
|
|
|
|
else if (fault & VM_FAULT_SIGBUS)
|
|
|
|
goto do_sigbus;
|
|
|
|
BUG();
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
2009-06-24 21:49:03 +04:00
|
|
|
|
|
|
|
if (fault & VM_FAULT_MAJOR) {
|
2007-07-19 12:47:05 +04:00
|
|
|
tsk->maj_flt++;
|
perf: Do the big rename: Performance Counters -> Performance Events
Bye-bye Performance Counters, welcome Performance Events!
In the past few months the perfcounters subsystem has grown out its
initial role of counting hardware events, and has become (and is
becoming) a much broader generic event enumeration, reporting, logging,
monitoring, analysis facility.
Naming its core object 'perf_counter' and naming the subsystem
'perfcounters' has become more and more of a misnomer. With pending
code like hw-breakpoints support the 'counter' name is less and
less appropriate.
All in one, we've decided to rename the subsystem to 'performance
events' and to propagate this rename through all fields, variables
and API names. (in an ABI compatible fashion)
The word 'event' is also a bit shorter than 'counter' - which makes
it slightly more convenient to write/handle as well.
Thanks goes to Stephane Eranian who first observed this misnomer and
suggested a rename.
User-space tooling and ABI compatibility is not affected - this patch
should be function-invariant. (Also, defconfigs were not touched to
keep the size down.)
This patch has been generated via the following script:
FILES=$(find * -type f | grep -vE 'oprofile|[^K]config')
sed -i \
-e 's/PERF_EVENT_/PERF_RECORD_/g' \
-e 's/PERF_COUNTER/PERF_EVENT/g' \
-e 's/perf_counter/perf_event/g' \
-e 's/nb_counters/nb_events/g' \
-e 's/swcounter/swevent/g' \
-e 's/tpcounter_event/tp_event/g' \
$FILES
for N in $(find . -name perf_counter.[ch]); do
M=$(echo $N | sed 's/perf_counter/perf_event/g')
mv $N $M
done
FILES=$(find . -name perf_event.*)
sed -i \
-e 's/COUNTER_MASK/REG_MASK/g' \
-e 's/COUNTER/EVENT/g' \
-e 's/\<event\>/event_id/g' \
-e 's/counter/event/g' \
-e 's/Counter/Event/g' \
$FILES
... to keep it as correct as possible. This script can also be
used by anyone who has pending perfcounters patches - it converts
a Linux kernel tree over to the new naming. We tried to time this
change to the point in time where the amount of pending patches
is the smallest: the end of the merge window.
Namespace clashes were fixed up in a preparatory patch - and some
stylistic fallout will be fixed up in a subsequent patch.
( NOTE: 'counters' are still the proper terminology when we deal
with hardware registers - and these sed scripts are a bit
over-eager in renaming them. I've undone some of that, but
in case there's something left where 'counter' would be
better than 'event' we can undo that on an individual basis
instead of touching an otherwise nicely automated patch. )
Suggested-by: Stephane Eranian <eranian@google.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Paul Mackerras <paulus@samba.org>
Reviewed-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <linux-arch@vger.kernel.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-09-21 14:02:48 +04:00
|
|
|
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1, 0,
|
2009-06-24 21:49:03 +04:00
|
|
|
regs, address);
|
|
|
|
} else {
|
2007-07-19 12:47:05 +04:00
|
|
|
tsk->min_flt++;
|
perf: Do the big rename: Performance Counters -> Performance Events
Bye-bye Performance Counters, welcome Performance Events!
In the past few months the perfcounters subsystem has grown out its
initial role of counting hardware events, and has become (and is
becoming) a much broader generic event enumeration, reporting, logging,
monitoring, analysis facility.
Naming its core object 'perf_counter' and naming the subsystem
'perfcounters' has become more and more of a misnomer. With pending
code like hw-breakpoints support the 'counter' name is less and
less appropriate.
All in one, we've decided to rename the subsystem to 'performance
events' and to propagate this rename through all fields, variables
and API names. (in an ABI compatible fashion)
The word 'event' is also a bit shorter than 'counter' - which makes
it slightly more convenient to write/handle as well.
Thanks goes to Stephane Eranian who first observed this misnomer and
suggested a rename.
User-space tooling and ABI compatibility is not affected - this patch
should be function-invariant. (Also, defconfigs were not touched to
keep the size down.)
This patch has been generated via the following script:
FILES=$(find * -type f | grep -vE 'oprofile|[^K]config')
sed -i \
-e 's/PERF_EVENT_/PERF_RECORD_/g' \
-e 's/PERF_COUNTER/PERF_EVENT/g' \
-e 's/perf_counter/perf_event/g' \
-e 's/nb_counters/nb_events/g' \
-e 's/swcounter/swevent/g' \
-e 's/tpcounter_event/tp_event/g' \
$FILES
for N in $(find . -name perf_counter.[ch]); do
M=$(echo $N | sed 's/perf_counter/perf_event/g')
mv $N $M
done
FILES=$(find . -name perf_event.*)
sed -i \
-e 's/COUNTER_MASK/REG_MASK/g' \
-e 's/COUNTER/EVENT/g' \
-e 's/\<event\>/event_id/g' \
-e 's/counter/event/g' \
-e 's/Counter/Event/g' \
$FILES
... to keep it as correct as possible. This script can also be
used by anyone who has pending perfcounters patches - it converts
a Linux kernel tree over to the new naming. We tried to time this
change to the point in time where the amount of pending patches
is the smallest: the end of the merge window.
Namespace clashes were fixed up in a preparatory patch - and some
stylistic fallout will be fixed up in a subsequent patch.
( NOTE: 'counters' are still the proper terminology when we deal
with hardware registers - and these sed scripts are a bit
over-eager in renaming them. I've undone some of that, but
in case there's something left where 'counter' would be
better than 'event' we can undo that on an individual basis
instead of touching an otherwise nicely automated patch. )
Suggested-by: Stephane Eranian <eranian@google.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Paul Mackerras <paulus@samba.org>
Reviewed-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: <linux-arch@vger.kernel.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-09-21 14:02:48 +04:00
|
|
|
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1, 0,
|
2009-06-24 21:49:03 +04:00
|
|
|
regs, address);
|
|
|
|
}
|
2007-07-19 12:47:05 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/* If we get here, the page fault has been handled. Do the TLB refill
|
|
|
|
now from the newly-setup PTE, to avoid having to fault again right
|
|
|
|
away on the same instruction. */
|
|
|
|
pte = lookup_pte (mm, address);
|
|
|
|
if (!pte) {
|
|
|
|
/* From empirical evidence, we can get here, due to
|
|
|
|
!pte_present(pte). (e.g. if a swap-in occurs, and the page
|
|
|
|
is swapped back out again before the process that wanted it
|
|
|
|
gets rescheduled?) */
|
|
|
|
goto no_pte;
|
|
|
|
}
|
|
|
|
|
|
|
|
__do_tlb_refill(address, textaccess, pte);
|
|
|
|
|
|
|
|
no_pte:
|
|
|
|
|
|
|
|
up_read(&mm->mmap_sem);
|
|
|
|
return;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Something tried to access memory that isn't in our memory map..
|
|
|
|
* Fix it, but check if it's kernel or user first..
|
|
|
|
*/
|
|
|
|
bad_area:
|
|
|
|
#ifdef DEBUG_FAULT
|
|
|
|
printk("fault:bad area\n");
|
|
|
|
#endif
|
|
|
|
up_read(&mm->mmap_sem);
|
|
|
|
|
|
|
|
if (user_mode(regs)) {
|
|
|
|
static int count=0;
|
|
|
|
siginfo_t info;
|
|
|
|
if (count < 4) {
|
|
|
|
/* This is really to help debug faults when starting
|
|
|
|
* usermode, so only need a few */
|
|
|
|
count++;
|
|
|
|
printk("user mode bad_area address=%08lx pid=%d (%s) pc=%08lx\n",
|
2007-10-19 10:40:41 +04:00
|
|
|
address, task_pid_nr(current), current->comm,
|
2005-04-17 02:20:36 +04:00
|
|
|
(unsigned long) regs->pc);
|
|
|
|
#if 0
|
|
|
|
show_regs(regs);
|
|
|
|
#endif
|
|
|
|
}
|
2007-10-19 10:39:52 +04:00
|
|
|
if (is_global_init(tsk)) {
|
2005-04-17 02:20:36 +04:00
|
|
|
panic("INIT had user mode bad_area\n");
|
|
|
|
}
|
|
|
|
tsk->thread.address = address;
|
|
|
|
tsk->thread.error_code = writeaccess;
|
|
|
|
info.si_signo = SIGSEGV;
|
|
|
|
info.si_errno = 0;
|
|
|
|
info.si_addr = (void *) address;
|
|
|
|
force_sig_info(SIGSEGV, &info, tsk);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
no_context:
|
|
|
|
#ifdef DEBUG_FAULT
|
|
|
|
printk("fault:No context\n");
|
|
|
|
#endif
|
|
|
|
/* Are we prepared to handle this kernel fault? */
|
|
|
|
fixup = search_exception_tables(regs->pc);
|
|
|
|
if (fixup) {
|
|
|
|
regs->pc = fixup->fixup;
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Oops. The kernel tried to access some bad page. We'll have to
|
|
|
|
* terminate things with extreme prejudice.
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
if (address < PAGE_SIZE)
|
|
|
|
printk(KERN_ALERT "Unable to handle kernel NULL pointer dereference");
|
|
|
|
else
|
|
|
|
printk(KERN_ALERT "Unable to handle kernel paging request");
|
|
|
|
printk(" at virtual address %08lx\n", address);
|
|
|
|
printk(KERN_ALERT "pc = %08Lx%08Lx\n", regs->pc >> 32, regs->pc & 0xffffffff);
|
|
|
|
die("Oops", regs, writeaccess);
|
|
|
|
do_exit(SIGKILL);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We ran out of memory, or some other thing happened to us that made
|
|
|
|
* us unable to handle the page fault gracefully.
|
|
|
|
*/
|
|
|
|
out_of_memory:
|
|
|
|
up_read(&mm->mmap_sem);
|
2010-04-22 20:06:26 +04:00
|
|
|
if (!user_mode(regs))
|
|
|
|
goto no_context;
|
|
|
|
pagefault_out_of_memory();
|
|
|
|
return;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
do_sigbus:
|
|
|
|
printk("fault:Do sigbus\n");
|
|
|
|
up_read(&mm->mmap_sem);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Send a sigbus, regardless of whether we were in kernel
|
|
|
|
* or user mode.
|
|
|
|
*/
|
|
|
|
tsk->thread.address = address;
|
|
|
|
tsk->thread.error_code = writeaccess;
|
|
|
|
tsk->thread.trap_no = 14;
|
|
|
|
force_sig(SIGBUS, tsk);
|
|
|
|
|
|
|
|
/* Kernel mode? Handle exceptions or die */
|
|
|
|
if (!user_mode(regs))
|
|
|
|
goto no_context;
|
|
|
|
}
|
|
|
|
|
2007-11-19 12:57:03 +03:00
|
|
|
void local_flush_tlb_one(unsigned long asid, unsigned long page)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
unsigned long long match, pteh=0, lpage;
|
|
|
|
unsigned long tlb;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Sign-extend based on neff.
|
|
|
|
*/
|
2009-08-04 12:14:39 +04:00
|
|
|
lpage = neff_sign_extend(page);
|
2007-11-19 12:57:03 +03:00
|
|
|
match = (asid << PTEH_ASID_SHIFT) | PTEH_VALID;
|
2005-04-17 02:20:36 +04:00
|
|
|
match |= lpage;
|
|
|
|
|
2007-11-19 12:57:03 +03:00
|
|
|
for_each_itlb_entry(tlb) {
|
|
|
|
asm volatile ("getcfg %1, 0, %0"
|
|
|
|
: "=r" (pteh)
|
|
|
|
: "r" (tlb) );
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2007-11-19 12:57:03 +03:00
|
|
|
if (pteh == match) {
|
|
|
|
__flush_tlb_slot(tlb);
|
|
|
|
break;
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
for_each_dtlb_entry(tlb) {
|
|
|
|
asm volatile ("getcfg %1, 0, %0"
|
|
|
|
: "=r" (pteh)
|
|
|
|
: "r" (tlb) );
|
|
|
|
|
|
|
|
if (pteh == match) {
|
|
|
|
__flush_tlb_slot(tlb);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2007-11-19 12:57:03 +03:00
|
|
|
void local_flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
|
|
|
|
if (vma->vm_mm) {
|
|
|
|
page &= PAGE_MASK;
|
|
|
|
local_irq_save(flags);
|
2007-11-19 12:57:03 +03:00
|
|
|
local_flush_tlb_one(get_asid(), page);
|
2005-04-17 02:20:36 +04:00
|
|
|
local_irq_restore(flags);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2007-11-19 12:57:03 +03:00
|
|
|
void local_flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
|
|
|
|
unsigned long end)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
unsigned long long match, pteh=0, pteh_epn, pteh_low;
|
|
|
|
unsigned long tlb;
|
2007-11-19 12:57:03 +03:00
|
|
|
unsigned int cpu = smp_processor_id();
|
2005-04-17 02:20:36 +04:00
|
|
|
struct mm_struct *mm;
|
|
|
|
|
|
|
|
mm = vma->vm_mm;
|
2007-11-19 12:57:03 +03:00
|
|
|
if (cpu_context(cpu, mm) == NO_CONTEXT)
|
2005-04-17 02:20:36 +04:00
|
|
|
return;
|
|
|
|
|
|
|
|
local_irq_save(flags);
|
|
|
|
|
|
|
|
start &= PAGE_MASK;
|
|
|
|
end &= PAGE_MASK;
|
|
|
|
|
2007-11-19 12:57:03 +03:00
|
|
|
match = (cpu_asid(cpu, mm) << PTEH_ASID_SHIFT) | PTEH_VALID;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
/* Flush ITLB */
|
|
|
|
for_each_itlb_entry(tlb) {
|
|
|
|
asm volatile ("getcfg %1, 0, %0"
|
|
|
|
: "=r" (pteh)
|
|
|
|
: "r" (tlb) );
|
|
|
|
|
|
|
|
pteh_epn = pteh & PAGE_MASK;
|
|
|
|
pteh_low = pteh & ~PAGE_MASK;
|
|
|
|
|
|
|
|
if (pteh_low == match && pteh_epn >= start && pteh_epn <= end)
|
|
|
|
__flush_tlb_slot(tlb);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Flush DTLB */
|
|
|
|
for_each_dtlb_entry(tlb) {
|
|
|
|
asm volatile ("getcfg %1, 0, %0"
|
|
|
|
: "=r" (pteh)
|
|
|
|
: "r" (tlb) );
|
|
|
|
|
|
|
|
pteh_epn = pteh & PAGE_MASK;
|
|
|
|
pteh_low = pteh & ~PAGE_MASK;
|
|
|
|
|
|
|
|
if (pteh_low == match && pteh_epn >= start && pteh_epn <= end)
|
|
|
|
__flush_tlb_slot(tlb);
|
|
|
|
}
|
|
|
|
|
|
|
|
local_irq_restore(flags);
|
|
|
|
}
|
|
|
|
|
2007-11-19 12:57:03 +03:00
|
|
|
void local_flush_tlb_mm(struct mm_struct *mm)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
unsigned long flags;
|
2007-11-19 12:57:03 +03:00
|
|
|
unsigned int cpu = smp_processor_id();
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2007-11-19 12:57:03 +03:00
|
|
|
if (cpu_context(cpu, mm) == NO_CONTEXT)
|
2005-04-17 02:20:36 +04:00
|
|
|
return;
|
|
|
|
|
|
|
|
local_irq_save(flags);
|
|
|
|
|
2007-11-19 12:57:03 +03:00
|
|
|
cpu_context(cpu, mm) = NO_CONTEXT;
|
|
|
|
if (mm == current->mm)
|
|
|
|
activate_context(mm, cpu);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
local_irq_restore(flags);
|
|
|
|
}
|
|
|
|
|
2007-11-19 12:57:03 +03:00
|
|
|
void local_flush_tlb_all(void)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
/* Invalidate all, including shared pages, excluding fixed TLBs */
|
|
|
|
unsigned long flags, tlb;
|
|
|
|
|
|
|
|
local_irq_save(flags);
|
|
|
|
|
|
|
|
/* Flush each ITLB entry */
|
2007-11-19 12:57:03 +03:00
|
|
|
for_each_itlb_entry(tlb)
|
2005-04-17 02:20:36 +04:00
|
|
|
__flush_tlb_slot(tlb);
|
|
|
|
|
|
|
|
/* Flush each DTLB entry */
|
2007-11-19 12:57:03 +03:00
|
|
|
for_each_dtlb_entry(tlb)
|
2005-04-17 02:20:36 +04:00
|
|
|
__flush_tlb_slot(tlb);
|
|
|
|
|
|
|
|
local_irq_restore(flags);
|
|
|
|
}
|
|
|
|
|
2007-11-19 12:57:03 +03:00
|
|
|
void local_flush_tlb_kernel_range(unsigned long start, unsigned long end)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
/* FIXME: Optimize this later.. */
|
|
|
|
flush_tlb_all();
|
|
|
|
}
|
2009-07-28 19:12:17 +04:00
|
|
|
|
|
|
|
void __update_tlb(struct vm_area_struct *vma, unsigned long address, pte_t pte)
|
|
|
|
{
|
|
|
|
}
|