2005-04-17 00:20:36 +02:00
|
|
|
/*
|
|
|
|
* arch/s390/mm/fault.c
|
|
|
|
*
|
|
|
|
* S390 version
|
|
|
|
* Copyright (C) 1999 IBM Deutschland Entwicklung GmbH, IBM Corporation
|
|
|
|
* Author(s): Hartmut Penner (hp@de.ibm.com)
|
|
|
|
* Ulrich Weigand (uweigand@de.ibm.com)
|
|
|
|
*
|
|
|
|
* Derived from "arch/i386/mm/fault.c"
|
|
|
|
* Copyright (C) 1995 Linus Torvalds
|
|
|
|
*/
|
|
|
|
|
2011-01-05 12:47:28 +01:00
|
|
|
#include <linux/kernel_stat.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 12:02:48 +02:00
|
|
|
#include <linux/perf_event.h>
|
2005-04-17 00:20:36 +02:00
|
|
|
#include <linux/signal.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>
|
2009-06-12 10:26:25 +02:00
|
|
|
#include <linux/compat.h>
|
2005-04-17 00:20:36 +02:00
|
|
|
#include <linux/smp.h>
|
2007-05-08 09:27:03 +02:00
|
|
|
#include <linux/kdebug.h>
|
2005-04-17 00:20:36 +02:00
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/console.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/hardirq.h>
|
2006-09-20 15:58:39 +02:00
|
|
|
#include <linux/kprobes.h>
|
2007-04-27 16:01:44 +02:00
|
|
|
#include <linux/uaccess.h>
|
2008-04-30 13:38:46 +02:00
|
|
|
#include <linux/hugetlb.h>
|
2010-02-26 22:37:43 +01:00
|
|
|
#include <asm/asm-offsets.h>
|
2005-04-17 00:20:36 +02:00
|
|
|
#include <asm/system.h>
|
|
|
|
#include <asm/pgtable.h>
|
2011-05-26 09:48:24 +02:00
|
|
|
#include <asm/irq.h>
|
2008-02-09 18:24:37 +01:00
|
|
|
#include <asm/mmu_context.h>
|
2009-12-07 12:51:45 +01:00
|
|
|
#include <asm/compat.h>
|
2008-04-17 07:46:26 +02:00
|
|
|
#include "../kernel/entry.h"
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2006-01-06 09:19:28 +01:00
|
|
|
#ifndef CONFIG_64BIT
|
2005-04-17 00:20:36 +02:00
|
|
|
#define __FAIL_ADDR_MASK 0x7ffff000
|
|
|
|
#define __SUBCODE_MASK 0x0200
|
|
|
|
#define __PF_RES_FIELD 0ULL
|
2006-01-06 09:19:28 +01:00
|
|
|
#else /* CONFIG_64BIT */
|
2005-04-17 00:20:36 +02:00
|
|
|
#define __FAIL_ADDR_MASK -4096L
|
|
|
|
#define __SUBCODE_MASK 0x0600
|
|
|
|
#define __PF_RES_FIELD 0x8000000000000000ULL
|
2006-01-06 09:19:28 +01:00
|
|
|
#endif /* CONFIG_64BIT */
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2009-12-07 12:51:45 +01:00
|
|
|
#define VM_FAULT_BADCONTEXT 0x010000
|
|
|
|
#define VM_FAULT_BADMAP 0x020000
|
|
|
|
#define VM_FAULT_BADACCESS 0x040000
|
|
|
|
|
2010-10-25 16:10:13 +02:00
|
|
|
static unsigned long store_indication;
|
|
|
|
|
|
|
|
void fault_init(void)
|
|
|
|
{
|
2010-10-25 16:10:51 +02:00
|
|
|
if (test_facility(2) && test_facility(75))
|
2010-10-25 16:10:13 +02:00
|
|
|
store_indication = 0xc00;
|
|
|
|
}
|
|
|
|
|
2009-12-07 12:51:44 +01:00
|
|
|
static inline int notify_page_fault(struct pt_regs *regs)
|
2007-04-27 16:01:43 +02:00
|
|
|
{
|
2007-05-04 18:47:46 +02:00
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
/* kprobe_running() needs smp_processor_id() */
|
2010-02-26 22:37:45 +01:00
|
|
|
if (kprobes_built_in() && !user_mode(regs)) {
|
2007-05-04 18:47:46 +02:00
|
|
|
preempt_disable();
|
|
|
|
if (kprobe_running() && kprobe_fault_handler(regs, 14))
|
|
|
|
ret = 1;
|
|
|
|
preempt_enable();
|
|
|
|
}
|
|
|
|
return ret;
|
2006-09-20 15:58:39 +02:00
|
|
|
}
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Unlock any spinlocks which will prevent us from getting the
|
2007-02-10 10:46:18 +01:00
|
|
|
* message out.
|
2005-04-17 00:20:36 +02:00
|
|
|
*/
|
|
|
|
void bust_spinlocks(int yes)
|
|
|
|
{
|
|
|
|
if (yes) {
|
|
|
|
oops_in_progress = 1;
|
|
|
|
} else {
|
|
|
|
int loglevel_save = console_loglevel;
|
|
|
|
console_unblank();
|
|
|
|
oops_in_progress = 0;
|
|
|
|
/*
|
|
|
|
* OK, the message is on the console. Now we call printk()
|
|
|
|
* without oops_in_progress set so that printk will give klogd
|
|
|
|
* a poke. Hold onto your hats...
|
|
|
|
*/
|
|
|
|
console_loglevel = 15;
|
|
|
|
printk(" ");
|
|
|
|
console_loglevel = loglevel_save;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2007-03-05 23:35:54 +01:00
|
|
|
* Returns the address space associated with the fault.
|
2009-12-07 12:51:42 +01:00
|
|
|
* Returns 0 for kernel space and 1 for user space.
|
2005-04-17 00:20:36 +02:00
|
|
|
*/
|
2009-12-07 12:51:42 +01:00
|
|
|
static inline int user_space_fault(unsigned long trans_exc_code)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
|
|
|
/*
|
2009-12-07 12:51:42 +01:00
|
|
|
* The lowest two bits of the translation exception
|
|
|
|
* identification indicate which paging table was used.
|
2005-04-17 00:20:36 +02:00
|
|
|
*/
|
2009-12-07 12:51:42 +01:00
|
|
|
trans_exc_code &= 3;
|
|
|
|
if (trans_exc_code == 2)
|
|
|
|
/* Access via secondary space, set_fs setting decides */
|
|
|
|
return current->thread.mm_segment.ar4;
|
2009-12-07 12:51:43 +01:00
|
|
|
if (user_mode == HOME_SPACE_MODE)
|
2009-12-07 12:51:42 +01:00
|
|
|
/* User space if the access has been done via home space. */
|
|
|
|
return trans_exc_code == 3;
|
|
|
|
/*
|
|
|
|
* If the user space is not the home space the kernel runs in home
|
|
|
|
* space. Access via secondary space has already been covered,
|
|
|
|
* access via primary space or access register is from user space
|
|
|
|
* and access via home space is from the kernel.
|
|
|
|
*/
|
|
|
|
return trans_exc_code != 3;
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
2010-05-17 10:00:21 +02:00
|
|
|
static inline void report_user_fault(struct pt_regs *regs, long int_code,
|
|
|
|
int signr, unsigned long address)
|
|
|
|
{
|
|
|
|
if ((task_pid_nr(current) > 1) && !show_unhandled_signals)
|
|
|
|
return;
|
|
|
|
if (!unhandled_signal(current, signr))
|
|
|
|
return;
|
|
|
|
if (!printk_ratelimit())
|
|
|
|
return;
|
|
|
|
printk("User process fault: interruption code 0x%lX ", int_code);
|
|
|
|
print_vma_addr(KERN_CONT "in ", regs->psw.addr & PSW_ADDR_INSN);
|
|
|
|
printk("\n");
|
|
|
|
printk("failing address: %lX\n", address);
|
|
|
|
show_regs(regs);
|
|
|
|
}
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
/*
|
|
|
|
* Send SIGSEGV to task. This is an external routine
|
|
|
|
* to keep the stack usage of do_page_fault small.
|
|
|
|
*/
|
2009-12-07 12:51:45 +01:00
|
|
|
static noinline void do_sigsegv(struct pt_regs *regs, long int_code,
|
|
|
|
int si_code, unsigned long trans_exc_code)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
|
|
|
struct siginfo si;
|
2009-12-07 12:51:45 +01:00
|
|
|
unsigned long address;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2009-12-07 12:51:45 +01:00
|
|
|
address = trans_exc_code & __FAIL_ADDR_MASK;
|
|
|
|
current->thread.prot_addr = address;
|
|
|
|
current->thread.trap_no = int_code;
|
2010-05-17 10:00:21 +02:00
|
|
|
report_user_fault(regs, int_code, SIGSEGV, address);
|
2005-04-17 00:20:36 +02:00
|
|
|
si.si_signo = SIGSEGV;
|
|
|
|
si.si_code = si_code;
|
2006-07-12 16:41:55 +02:00
|
|
|
si.si_addr = (void __user *) address;
|
2005-04-17 00:20:36 +02:00
|
|
|
force_sig_info(SIGSEGV, &si, current);
|
|
|
|
}
|
|
|
|
|
2009-12-07 12:51:45 +01:00
|
|
|
static noinline void do_no_context(struct pt_regs *regs, long int_code,
|
|
|
|
unsigned long trans_exc_code)
|
2007-04-27 16:01:43 +02:00
|
|
|
{
|
|
|
|
const struct exception_table_entry *fixup;
|
2009-12-07 12:51:42 +01:00
|
|
|
unsigned long address;
|
2007-04-27 16:01:43 +02:00
|
|
|
|
|
|
|
/* Are we prepared to handle this kernel fault? */
|
2009-12-07 12:51:45 +01:00
|
|
|
fixup = search_exception_tables(regs->psw.addr & PSW_ADDR_INSN);
|
2007-04-27 16:01:43 +02:00
|
|
|
if (fixup) {
|
|
|
|
regs->psw.addr = fixup->fixup | PSW_ADDR_AMODE;
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Oops. The kernel tried to access some bad page. We'll have to
|
|
|
|
* terminate things with extreme prejudice.
|
|
|
|
*/
|
2009-12-07 12:51:42 +01:00
|
|
|
address = trans_exc_code & __FAIL_ADDR_MASK;
|
2009-12-07 12:51:43 +01:00
|
|
|
if (!user_space_fault(trans_exc_code))
|
2007-04-27 16:01:43 +02:00
|
|
|
printk(KERN_ALERT "Unable to handle kernel pointer dereference"
|
|
|
|
" at virtual kernel address %p\n", (void *)address);
|
|
|
|
else
|
|
|
|
printk(KERN_ALERT "Unable to handle kernel paging request"
|
|
|
|
" at virtual user address %p\n", (void *)address);
|
|
|
|
|
2009-12-07 12:51:45 +01:00
|
|
|
die("Oops", regs, int_code);
|
2007-04-27 16:01:43 +02:00
|
|
|
do_exit(SIGKILL);
|
|
|
|
}
|
|
|
|
|
2009-12-07 12:51:45 +01:00
|
|
|
static noinline void do_low_address(struct pt_regs *regs, long int_code,
|
|
|
|
unsigned long trans_exc_code)
|
2007-04-27 16:01:43 +02:00
|
|
|
{
|
|
|
|
/* Low-address protection hit in kernel mode means
|
|
|
|
NULL pointer write access in kernel mode. */
|
|
|
|
if (regs->psw.mask & PSW_MASK_PSTATE) {
|
|
|
|
/* Low-address protection hit in user mode 'cannot happen'. */
|
2009-12-07 12:51:45 +01:00
|
|
|
die ("Low-address protection", regs, int_code);
|
2007-04-27 16:01:43 +02:00
|
|
|
do_exit(SIGKILL);
|
|
|
|
}
|
|
|
|
|
2009-12-07 12:51:45 +01:00
|
|
|
do_no_context(regs, int_code, trans_exc_code);
|
2007-04-27 16:01:43 +02:00
|
|
|
}
|
|
|
|
|
2009-12-07 12:51:45 +01:00
|
|
|
static noinline void do_sigbus(struct pt_regs *regs, long int_code,
|
|
|
|
unsigned long trans_exc_code)
|
2007-04-27 16:01:43 +02:00
|
|
|
{
|
|
|
|
struct task_struct *tsk = current;
|
2010-10-25 16:10:35 +02:00
|
|
|
unsigned long address;
|
|
|
|
struct siginfo si;
|
2007-04-27 16:01:43 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Send a sigbus, regardless of whether we were in kernel
|
|
|
|
* or user mode.
|
|
|
|
*/
|
2010-10-25 16:10:35 +02:00
|
|
|
address = trans_exc_code & __FAIL_ADDR_MASK;
|
|
|
|
tsk->thread.prot_addr = address;
|
2009-12-07 12:51:45 +01:00
|
|
|
tsk->thread.trap_no = int_code;
|
2010-10-25 16:10:35 +02:00
|
|
|
si.si_signo = SIGBUS;
|
|
|
|
si.si_errno = 0;
|
|
|
|
si.si_code = BUS_ADRERR;
|
|
|
|
si.si_addr = (void __user *) address;
|
|
|
|
force_sig_info(SIGBUS, &si, tsk);
|
2007-04-27 16:01:43 +02:00
|
|
|
}
|
|
|
|
|
2009-12-07 12:51:45 +01:00
|
|
|
static noinline void do_fault_error(struct pt_regs *regs, long int_code,
|
|
|
|
unsigned long trans_exc_code, int fault)
|
|
|
|
{
|
|
|
|
int si_code;
|
|
|
|
|
|
|
|
switch (fault) {
|
|
|
|
case VM_FAULT_BADACCESS:
|
|
|
|
case VM_FAULT_BADMAP:
|
|
|
|
/* Bad memory access. Check if it is kernel or user space. */
|
|
|
|
if (regs->psw.mask & PSW_MASK_PSTATE) {
|
|
|
|
/* User mode accesses just cause a SIGSEGV */
|
|
|
|
si_code = (fault == VM_FAULT_BADMAP) ?
|
|
|
|
SEGV_MAPERR : SEGV_ACCERR;
|
|
|
|
do_sigsegv(regs, int_code, si_code, trans_exc_code);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
case VM_FAULT_BADCONTEXT:
|
|
|
|
do_no_context(regs, int_code, trans_exc_code);
|
|
|
|
break;
|
|
|
|
default: /* fault & VM_FAULT_ERROR */
|
2011-05-26 09:48:29 +02:00
|
|
|
if (fault & VM_FAULT_OOM) {
|
|
|
|
if (!(regs->psw.mask & PSW_MASK_PSTATE))
|
|
|
|
do_no_context(regs, int_code, trans_exc_code);
|
|
|
|
else
|
|
|
|
pagefault_out_of_memory();
|
|
|
|
} else if (fault & VM_FAULT_SIGBUS) {
|
2009-12-07 12:51:45 +01:00
|
|
|
/* Kernel mode? Handle exceptions or die */
|
|
|
|
if (!(regs->psw.mask & PSW_MASK_PSTATE))
|
|
|
|
do_no_context(regs, int_code, trans_exc_code);
|
2010-10-25 16:10:35 +02:00
|
|
|
else
|
|
|
|
do_sigbus(regs, int_code, trans_exc_code);
|
2009-12-07 12:51:45 +01:00
|
|
|
} else
|
|
|
|
BUG();
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
/*
|
|
|
|
* This routine handles page faults. It determines the address,
|
|
|
|
* and the problem, and then passes it off to one of the appropriate
|
|
|
|
* routines.
|
|
|
|
*
|
2009-12-07 12:51:45 +01:00
|
|
|
* interruption code (int_code):
|
2005-04-17 00:20:36 +02:00
|
|
|
* 04 Protection -> Write-Protection (suprression)
|
|
|
|
* 10 Segment translation -> Not present (nullification)
|
|
|
|
* 11 Page translation -> Not present (nullification)
|
|
|
|
* 3b Region third trans. -> Not present (nullification)
|
|
|
|
*/
|
2009-12-07 12:51:46 +01:00
|
|
|
static inline int do_exception(struct pt_regs *regs, int access,
|
2009-12-07 12:51:45 +01:00
|
|
|
unsigned long trans_exc_code)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
2007-04-27 16:01:43 +02:00
|
|
|
struct task_struct *tsk;
|
|
|
|
struct mm_struct *mm;
|
|
|
|
struct vm_area_struct *vma;
|
|
|
|
unsigned long address;
|
2011-05-26 09:48:30 +02:00
|
|
|
unsigned int flags;
|
|
|
|
int fault;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2009-12-07 12:51:44 +01:00
|
|
|
if (notify_page_fault(regs))
|
2009-12-07 12:51:45 +01:00
|
|
|
return 0;
|
2006-09-20 15:58:39 +02:00
|
|
|
|
2007-04-27 16:01:43 +02:00
|
|
|
tsk = current;
|
|
|
|
mm = tsk->mm;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Verify that the fault happened in user space, that
|
|
|
|
* we are not in an interrupt and that there is a
|
|
|
|
* user context.
|
|
|
|
*/
|
2009-12-07 12:51:45 +01:00
|
|
|
fault = VM_FAULT_BADCONTEXT;
|
2009-12-07 12:51:42 +01:00
|
|
|
if (unlikely(!user_space_fault(trans_exc_code) || in_atomic() || !mm))
|
2009-12-07 12:51:45 +01:00
|
|
|
goto out;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2009-12-07 12:51:42 +01:00
|
|
|
address = trans_exc_code & __FAIL_ADDR_MASK;
|
2011-06-27 14:41:57 +02:00
|
|
|
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
|
2011-05-26 09:48:30 +02:00
|
|
|
flags = FAULT_FLAG_ALLOW_RETRY;
|
|
|
|
if (access == VM_WRITE || (trans_exc_code & store_indication) == 0x400)
|
|
|
|
flags |= FAULT_FLAG_WRITE;
|
2007-04-27 16:01:43 +02:00
|
|
|
down_read(&mm->mmap_sem);
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2011-07-24 10:48:20 +02:00
|
|
|
#ifdef CONFIG_PGSTE
|
|
|
|
if (test_tsk_thread_flag(current, TIF_SIE) && S390_lowcore.gmap) {
|
|
|
|
address = gmap_fault(address,
|
|
|
|
(struct gmap *) S390_lowcore.gmap);
|
|
|
|
if (address == -EFAULT) {
|
|
|
|
fault = VM_FAULT_BADMAP;
|
|
|
|
goto out_up;
|
|
|
|
}
|
|
|
|
if (address == -ENOMEM) {
|
|
|
|
fault = VM_FAULT_OOM;
|
|
|
|
goto out_up;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
|
|
|
retry:
|
2009-12-07 12:51:45 +01:00
|
|
|
fault = VM_FAULT_BADMAP;
|
2007-03-05 23:35:54 +01:00
|
|
|
vma = find_vma(mm, address);
|
|
|
|
if (!vma)
|
2009-12-07 12:51:45 +01:00
|
|
|
goto out_up;
|
2007-02-05 21:18:17 +01:00
|
|
|
|
2009-12-07 12:51:45 +01:00
|
|
|
if (unlikely(vma->vm_start > address)) {
|
|
|
|
if (!(vma->vm_flags & VM_GROWSDOWN))
|
|
|
|
goto out_up;
|
|
|
|
if (expand_stack(vma, address))
|
|
|
|
goto out_up;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Ok, we have a good vm_area for this memory access, so
|
|
|
|
* we can handle it..
|
|
|
|
*/
|
|
|
|
fault = VM_FAULT_BADACCESS;
|
2009-12-07 12:51:46 +01:00
|
|
|
if (unlikely(!(vma->vm_flags & access)))
|
2009-12-07 12:51:45 +01:00
|
|
|
goto out_up;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2008-04-30 13:38:46 +02:00
|
|
|
if (is_vm_hugetlb_page(vma))
|
|
|
|
address &= HPAGE_MASK;
|
2005-04-17 00:20:36 +02:00
|
|
|
/*
|
|
|
|
* If for any reason at all we couldn't handle the fault,
|
|
|
|
* make sure we exit gracefully rather than endlessly redo
|
|
|
|
* the fault.
|
|
|
|
*/
|
2011-05-26 09:48:30 +02:00
|
|
|
fault = handle_mm_fault(mm, vma, address, flags);
|
2009-12-07 12:51:45 +01:00
|
|
|
if (unlikely(fault & VM_FAULT_ERROR))
|
|
|
|
goto out_up;
|
|
|
|
|
2011-05-26 09:48:30 +02:00
|
|
|
/*
|
|
|
|
* Major/minor page fault accounting is only done on the
|
|
|
|
* initial attempt. If we go through a retry, it is extremely
|
|
|
|
* likely that the page will be found in page cache at that point.
|
|
|
|
*/
|
|
|
|
if (flags & FAULT_FLAG_ALLOW_RETRY) {
|
|
|
|
if (fault & VM_FAULT_MAJOR) {
|
|
|
|
tsk->maj_flt++;
|
2011-06-27 14:41:57 +02:00
|
|
|
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1,
|
2011-05-26 09:48:30 +02:00
|
|
|
regs, address);
|
|
|
|
} else {
|
|
|
|
tsk->min_flt++;
|
2011-06-27 14:41:57 +02:00
|
|
|
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1,
|
2011-05-26 09:48:30 +02:00
|
|
|
regs, address);
|
|
|
|
}
|
|
|
|
if (fault & VM_FAULT_RETRY) {
|
|
|
|
/* Clear FAULT_FLAG_ALLOW_RETRY to avoid any risk
|
|
|
|
* of starvation. */
|
|
|
|
flags &= ~FAULT_FLAG_ALLOW_RETRY;
|
2011-07-24 10:48:20 +02:00
|
|
|
down_read(&mm->mmap_sem);
|
2011-05-26 09:48:30 +02:00
|
|
|
goto retry;
|
|
|
|
}
|
2009-09-11 10:29:06 +02:00
|
|
|
}
|
2005-04-17 00:20:36 +02:00
|
|
|
/*
|
|
|
|
* The instruction that caused the program check will
|
|
|
|
* be repeated. Don't signal single step via SIGTRAP.
|
|
|
|
*/
|
2011-01-05 12:48:10 +01:00
|
|
|
clear_tsk_thread_flag(tsk, TIF_PER_TRAP);
|
2009-12-07 12:51:45 +01:00
|
|
|
fault = 0;
|
|
|
|
out_up:
|
2007-04-27 16:01:43 +02:00
|
|
|
up_read(&mm->mmap_sem);
|
2009-12-07 12:51:45 +01:00
|
|
|
out:
|
|
|
|
return fault;
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
2010-10-25 16:10:37 +02:00
|
|
|
void __kprobes do_protection_exception(struct pt_regs *regs, long pgm_int_code,
|
|
|
|
unsigned long trans_exc_code)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
2009-12-07 12:51:45 +01:00
|
|
|
int fault;
|
2009-12-07 12:51:42 +01:00
|
|
|
|
2011-03-31 03:57:33 +02:00
|
|
|
/* Protection exception is suppressing, decrement psw address. */
|
2011-10-30 15:16:48 +01:00
|
|
|
regs->psw.addr = __rewind_psw(regs->psw, pgm_int_code >> 16);
|
2007-04-27 16:01:43 +02:00
|
|
|
/*
|
|
|
|
* Check for low-address protection. This needs to be treated
|
|
|
|
* as a special case because the translation exception code
|
|
|
|
* field is not guaranteed to contain valid data in this case.
|
|
|
|
*/
|
2009-12-07 12:51:42 +01:00
|
|
|
if (unlikely(!(trans_exc_code & 4))) {
|
2010-10-25 16:10:37 +02:00
|
|
|
do_low_address(regs, pgm_int_code, trans_exc_code);
|
2007-04-27 16:01:43 +02:00
|
|
|
return;
|
|
|
|
}
|
2009-12-07 12:51:46 +01:00
|
|
|
fault = do_exception(regs, VM_WRITE, trans_exc_code);
|
2009-12-07 12:51:45 +01:00
|
|
|
if (unlikely(fault))
|
|
|
|
do_fault_error(regs, 4, trans_exc_code, fault);
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
2010-10-25 16:10:37 +02:00
|
|
|
void __kprobes do_dat_exception(struct pt_regs *regs, long pgm_int_code,
|
|
|
|
unsigned long trans_exc_code)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
2009-12-07 12:51:46 +01:00
|
|
|
int access, fault;
|
2009-12-07 12:51:45 +01:00
|
|
|
|
2009-12-07 12:51:46 +01:00
|
|
|
access = VM_READ | VM_EXEC | VM_WRITE;
|
|
|
|
fault = do_exception(regs, access, trans_exc_code);
|
2009-12-07 12:51:45 +01:00
|
|
|
if (unlikely(fault))
|
2010-10-25 16:10:37 +02:00
|
|
|
do_fault_error(regs, pgm_int_code & 255, trans_exc_code, fault);
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
2008-02-09 18:24:37 +01:00
|
|
|
#ifdef CONFIG_64BIT
|
2010-10-25 16:10:37 +02:00
|
|
|
void __kprobes do_asce_exception(struct pt_regs *regs, long pgm_int_code,
|
|
|
|
unsigned long trans_exc_code)
|
2008-02-09 18:24:37 +01:00
|
|
|
{
|
2009-12-07 12:51:45 +01:00
|
|
|
struct mm_struct *mm = current->mm;
|
2008-02-09 18:24:37 +01:00
|
|
|
struct vm_area_struct *vma;
|
|
|
|
|
2009-12-07 12:51:42 +01:00
|
|
|
if (unlikely(!user_space_fault(trans_exc_code) || in_atomic() || !mm))
|
2008-02-09 18:24:37 +01:00
|
|
|
goto no_context;
|
|
|
|
|
|
|
|
down_read(&mm->mmap_sem);
|
2009-12-07 12:51:45 +01:00
|
|
|
vma = find_vma(mm, trans_exc_code & __FAIL_ADDR_MASK);
|
2008-02-09 18:24:37 +01:00
|
|
|
up_read(&mm->mmap_sem);
|
|
|
|
|
|
|
|
if (vma) {
|
|
|
|
update_mm(mm, current);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* User mode accesses just cause a SIGSEGV */
|
|
|
|
if (regs->psw.mask & PSW_MASK_PSTATE) {
|
2010-10-25 16:10:37 +02:00
|
|
|
do_sigsegv(regs, pgm_int_code, SEGV_MAPERR, trans_exc_code);
|
2008-02-09 18:24:37 +01:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
no_context:
|
2010-10-25 16:10:37 +02:00
|
|
|
do_no_context(regs, pgm_int_code, trans_exc_code);
|
2008-02-09 18:24:37 +01:00
|
|
|
}
|
|
|
|
#endif
|
|
|
|
|
2010-10-25 16:10:37 +02:00
|
|
|
int __handle_fault(unsigned long uaddr, unsigned long pgm_int_code, int write)
|
2009-12-07 12:51:47 +01:00
|
|
|
{
|
|
|
|
struct pt_regs regs;
|
|
|
|
int access, fault;
|
|
|
|
|
|
|
|
regs.psw.mask = psw_kernel_bits;
|
|
|
|
if (!irqs_disabled())
|
|
|
|
regs.psw.mask |= PSW_MASK_IO | PSW_MASK_EXT;
|
|
|
|
regs.psw.addr = (unsigned long) __builtin_return_address(0);
|
|
|
|
regs.psw.addr |= PSW_ADDR_AMODE;
|
|
|
|
uaddr &= PAGE_MASK;
|
2010-10-25 16:10:37 +02:00
|
|
|
access = write ? VM_WRITE : VM_READ;
|
2009-12-07 12:51:47 +01:00
|
|
|
fault = do_exception(®s, access, uaddr | 2);
|
|
|
|
if (unlikely(fault)) {
|
2011-05-26 09:48:29 +02:00
|
|
|
if (fault & VM_FAULT_OOM)
|
|
|
|
return -EFAULT;
|
|
|
|
else if (fault & VM_FAULT_SIGBUS)
|
2010-10-25 16:10:37 +02:00
|
|
|
do_sigbus(®s, pgm_int_code, uaddr);
|
2009-12-07 12:51:47 +01:00
|
|
|
}
|
|
|
|
return fault ? -EFAULT : 0;
|
|
|
|
}
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
#ifdef CONFIG_PFAULT
|
|
|
|
/*
|
|
|
|
* 'pfault' pseudo page faults routines.
|
|
|
|
*/
|
2011-01-05 12:47:39 +01:00
|
|
|
static int pfault_disable;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
|
|
|
static int __init nopfault(char *str)
|
|
|
|
{
|
|
|
|
pfault_disable = 1;
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
__setup("nopfault", nopfault);
|
|
|
|
|
2011-05-23 10:24:35 +02:00
|
|
|
struct pfault_refbk {
|
|
|
|
u16 refdiagc;
|
|
|
|
u16 reffcode;
|
|
|
|
u16 refdwlen;
|
|
|
|
u16 refversn;
|
|
|
|
u64 refgaddr;
|
|
|
|
u64 refselmk;
|
|
|
|
u64 refcmpmk;
|
|
|
|
u64 reserved;
|
|
|
|
} __attribute__ ((packed, aligned(8)));
|
2005-04-17 00:20:36 +02:00
|
|
|
|
|
|
|
int pfault_init(void)
|
|
|
|
{
|
2011-05-23 10:24:35 +02:00
|
|
|
struct pfault_refbk refbk = {
|
|
|
|
.refdiagc = 0x258,
|
|
|
|
.reffcode = 0,
|
|
|
|
.refdwlen = 5,
|
|
|
|
.refversn = 2,
|
|
|
|
.refgaddr = __LC_CURRENT_PID,
|
|
|
|
.refselmk = 1ULL << 48,
|
|
|
|
.refcmpmk = 1ULL << 48,
|
|
|
|
.reserved = __PF_RES_FIELD };
|
2005-04-17 00:20:36 +02:00
|
|
|
int rc;
|
|
|
|
|
2006-12-04 15:40:40 +01:00
|
|
|
if (!MACHINE_IS_VM || pfault_disable)
|
2005-04-17 00:20:36 +02:00
|
|
|
return -1;
|
2006-09-28 16:56:43 +02:00
|
|
|
asm volatile(
|
|
|
|
" diag %1,%0,0x258\n"
|
|
|
|
"0: j 2f\n"
|
|
|
|
"1: la %0,8\n"
|
2005-04-17 00:20:36 +02:00
|
|
|
"2:\n"
|
2006-09-28 16:56:43 +02:00
|
|
|
EX_TABLE(0b,1b)
|
|
|
|
: "=d" (rc) : "a" (&refbk), "m" (refbk) : "cc");
|
2005-04-17 00:20:36 +02:00
|
|
|
return rc;
|
|
|
|
}
|
|
|
|
|
|
|
|
void pfault_fini(void)
|
|
|
|
{
|
2011-05-23 10:24:35 +02:00
|
|
|
struct pfault_refbk refbk = {
|
|
|
|
.refdiagc = 0x258,
|
|
|
|
.reffcode = 1,
|
|
|
|
.refdwlen = 5,
|
|
|
|
.refversn = 2,
|
|
|
|
};
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2006-12-04 15:40:40 +01:00
|
|
|
if (!MACHINE_IS_VM || pfault_disable)
|
2005-04-17 00:20:36 +02:00
|
|
|
return;
|
2006-09-28 16:56:43 +02:00
|
|
|
asm volatile(
|
|
|
|
" diag %0,0,0x258\n"
|
2005-04-17 00:20:36 +02:00
|
|
|
"0:\n"
|
2006-09-28 16:56:43 +02:00
|
|
|
EX_TABLE(0b,0b)
|
|
|
|
: : "a" (&refbk), "m" (refbk) : "cc");
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
2011-05-23 10:24:34 +02:00
|
|
|
static DEFINE_SPINLOCK(pfault_lock);
|
|
|
|
static LIST_HEAD(pfault_list);
|
|
|
|
|
2010-10-25 16:10:38 +02:00
|
|
|
static void pfault_interrupt(unsigned int ext_int_code,
|
|
|
|
unsigned int param32, unsigned long param64)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
|
|
|
struct task_struct *tsk;
|
|
|
|
__u16 subcode;
|
2011-05-23 10:24:34 +02:00
|
|
|
pid_t pid;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Get the external interruption subcode & pfault
|
|
|
|
* initial/completion signal bit. VM stores this
|
|
|
|
* in the 'cpu address' field associated with the
|
|
|
|
* external interrupt.
|
|
|
|
*/
|
2010-10-25 16:10:38 +02:00
|
|
|
subcode = ext_int_code >> 16;
|
2005-04-17 00:20:36 +02:00
|
|
|
if ((subcode & 0xff00) != __SUBCODE_MASK)
|
|
|
|
return;
|
2011-04-29 10:42:19 +02:00
|
|
|
kstat_cpu(smp_processor_id()).irqs[EXTINT_PFL]++;
|
2011-05-23 10:24:34 +02:00
|
|
|
if (subcode & 0x0080) {
|
|
|
|
/* Get the token (= pid of the affected task). */
|
|
|
|
pid = sizeof(void *) == 4 ? param32 : param64;
|
|
|
|
rcu_read_lock();
|
|
|
|
tsk = find_task_by_pid_ns(pid, &init_pid_ns);
|
|
|
|
if (tsk)
|
|
|
|
get_task_struct(tsk);
|
|
|
|
rcu_read_unlock();
|
|
|
|
if (!tsk)
|
|
|
|
return;
|
|
|
|
} else {
|
|
|
|
tsk = current;
|
|
|
|
}
|
|
|
|
spin_lock(&pfault_lock);
|
2005-04-17 00:20:36 +02:00
|
|
|
if (subcode & 0x0080) {
|
|
|
|
/* signal bit is set -> a page has been swapped in by VM */
|
2011-05-23 10:24:34 +02:00
|
|
|
if (tsk->thread.pfault_wait == 1) {
|
2005-04-17 00:20:36 +02:00
|
|
|
/* Initial interrupt was faster than the completion
|
|
|
|
* interrupt. pfault_wait is valid. Set pfault_wait
|
|
|
|
* back to zero and wake up the process. This can
|
|
|
|
* safely be done because the task is still sleeping
|
2005-09-04 00:58:02 +02:00
|
|
|
* and can't produce new pfaults. */
|
2005-04-17 00:20:36 +02:00
|
|
|
tsk->thread.pfault_wait = 0;
|
2011-05-23 10:24:34 +02:00
|
|
|
list_del(&tsk->thread.list);
|
2005-04-17 00:20:36 +02:00
|
|
|
wake_up_process(tsk);
|
2011-05-23 10:24:34 +02:00
|
|
|
} else {
|
|
|
|
/* Completion interrupt was faster than initial
|
|
|
|
* interrupt. Set pfault_wait to -1 so the initial
|
|
|
|
* interrupt doesn't put the task to sleep. */
|
|
|
|
tsk->thread.pfault_wait = -1;
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
2011-05-23 10:24:34 +02:00
|
|
|
put_task_struct(tsk);
|
2005-04-17 00:20:36 +02:00
|
|
|
} else {
|
|
|
|
/* signal bit not set -> a real page is missing. */
|
2011-05-23 10:24:34 +02:00
|
|
|
if (tsk->thread.pfault_wait == -1) {
|
2005-04-17 00:20:36 +02:00
|
|
|
/* Completion interrupt was faster than the initial
|
2011-05-23 10:24:34 +02:00
|
|
|
* interrupt (pfault_wait == -1). Set pfault_wait
|
|
|
|
* back to zero and exit. */
|
2005-04-17 00:20:36 +02:00
|
|
|
tsk->thread.pfault_wait = 0;
|
2011-05-23 10:24:34 +02:00
|
|
|
} else {
|
|
|
|
/* Initial interrupt arrived before completion
|
|
|
|
* interrupt. Let the task sleep. */
|
|
|
|
tsk->thread.pfault_wait = 1;
|
|
|
|
list_add(&tsk->thread.list, &pfault_list);
|
|
|
|
set_task_state(tsk, TASK_UNINTERRUPTIBLE);
|
2005-04-17 00:20:36 +02:00
|
|
|
set_tsk_need_resched(tsk);
|
2011-05-23 10:24:34 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
spin_unlock(&pfault_lock);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int __cpuinit pfault_cpu_notify(struct notifier_block *self,
|
|
|
|
unsigned long action, void *hcpu)
|
|
|
|
{
|
|
|
|
struct thread_struct *thread, *next;
|
|
|
|
struct task_struct *tsk;
|
|
|
|
|
|
|
|
switch (action) {
|
|
|
|
case CPU_DEAD:
|
|
|
|
case CPU_DEAD_FROZEN:
|
|
|
|
spin_lock_irq(&pfault_lock);
|
|
|
|
list_for_each_entry_safe(thread, next, &pfault_list, list) {
|
|
|
|
thread->pfault_wait = 0;
|
|
|
|
list_del(&thread->list);
|
|
|
|
tsk = container_of(thread, struct task_struct, thread);
|
|
|
|
wake_up_process(tsk);
|
|
|
|
}
|
|
|
|
spin_unlock_irq(&pfault_lock);
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
break;
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
2011-05-23 10:24:34 +02:00
|
|
|
return NOTIFY_OK;
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
2011-01-05 12:47:39 +01:00
|
|
|
static int __init pfault_irq_init(void)
|
2006-12-04 15:40:40 +01:00
|
|
|
{
|
2011-01-05 12:47:39 +01:00
|
|
|
int rc;
|
2006-12-04 15:40:40 +01:00
|
|
|
|
2011-01-05 12:47:39 +01:00
|
|
|
if (!MACHINE_IS_VM)
|
|
|
|
return 0;
|
|
|
|
rc = register_external_interrupt(0x2603, pfault_interrupt);
|
2011-05-23 10:24:35 +02:00
|
|
|
if (rc)
|
|
|
|
goto out_extint;
|
|
|
|
rc = pfault_init() == 0 ? 0 : -EOPNOTSUPP;
|
|
|
|
if (rc)
|
|
|
|
goto out_pfault;
|
2011-05-26 09:48:23 +02:00
|
|
|
service_subclass_irq_register();
|
2011-05-23 10:24:35 +02:00
|
|
|
hotcpu_notifier(pfault_cpu_notify, 0);
|
|
|
|
return 0;
|
2006-12-04 15:40:40 +01:00
|
|
|
|
2011-05-23 10:24:35 +02:00
|
|
|
out_pfault:
|
2011-01-05 12:47:39 +01:00
|
|
|
unregister_external_interrupt(0x2603, pfault_interrupt);
|
2011-05-23 10:24:35 +02:00
|
|
|
out_extint:
|
|
|
|
pfault_disable = 1;
|
|
|
|
return rc;
|
2006-12-04 15:40:40 +01:00
|
|
|
}
|
2011-01-05 12:47:39 +01:00
|
|
|
early_initcall(pfault_irq_init);
|
|
|
|
|
2011-05-23 10:24:35 +02:00
|
|
|
#endif /* CONFIG_PFAULT */
|