/*
* This routine handles page faults. It determines the address,
* and the problem, and then passes it off to one of the appropriate
* routines.
*
* interruption code (int_code):
* 04 Protection -> Write-Protection (suprression)
* 10 Segment translation -> Not present (nullification)
* 11 Page translation -> Not present (nullification)
* 3b Region third trans. -> Not present (nullification)
*/
static inline int do_exception(struct pt_regs *regs, int access)
{
struct task_struct *tsk;
struct mm_struct *mm;
struct vm_area_struct *vma;
unsigned long trans_exc_code;
unsigned long address;
unsigned int flags;
int fault;
if (notify_page_fault(regs))
return 0;
tsk = current;
mm = tsk->mm;
trans_exc_code = regs->int_parm_long;
/*
* Verify that the fault happened in user space, that
* we are not in an interrupt and that there is a
* user context.
*/
fault = VM_FAULT_BADCONTEXT;
if (unlikely(!user_space_fault(trans_exc_code) || in_atomic() || !mm))
goto out;
address = trans_exc_code & __FAIL_ADDR_MASK;
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
flags = FAULT_FLAG_ALLOW_RETRY;
if (access == VM_WRITE || (trans_exc_code & store_indication) == 0x400)
flags |= FAULT_FLAG_WRITE;
down_read(&mm->mmap_sem);
#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:
fault = VM_FAULT_BADMAP;
vma = find_vma(mm, address);
if (!vma)
goto out_up;
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;
if (unlikely(!(vma->vm_flags & access)))
goto out_up;
if (is_vm_hugetlb_page(vma))
address &= HPAGE_MASK;
/*
* If for any reason at all we couldn't handle the fault,
* make sure we exit gracefully rather than endlessly redo
* the fault.
*/
fault = handle_mm_fault(mm, vma, address, flags);
if (unlikely(fault & VM_FAULT_ERROR))
goto out_up;
/*
* 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++;
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1,
regs, address);
} else {
tsk->min_flt++;
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1,
regs, address);
}
if (fault & VM_FAULT_RETRY) {
/* Clear FAULT_FLAG_ALLOW_RETRY to avoid any risk
* of starvation. */
flags &= ~FAULT_FLAG_ALLOW_RETRY;
flags |= FAULT_FLAG_TRIED;
down_read(&mm->mmap_sem);
goto retry;
}
}
/*
* The instruction that caused the program check will
* be repeated. Don't signal single step via SIGTRAP.
*/
clear_tsk_thread_flag(tsk, TIF_PER_TRAP);
fault = 0;
out_up:
up_read(&mm->mmap_sem);
out:
return fault;
}
static inline int do_exception(struct pt_regs *regs, int access)
{
struct task_struct *tsk;
struct mm_struct *mm;
struct vm_area_struct *vma;
unsigned long trans_exc_code;
unsigned long address;
unsigned int flags;
int fault;
if (notify_page_fault(regs))
return 0;
tsk = current;
mm = tsk->mm;
trans_exc_code = regs->int_parm_long;
fault = VM_FAULT_BADCONTEXT;
if (unlikely(!user_space_fault(trans_exc_code) || in_atomic() || !mm))
goto out;
address = trans_exc_code & __FAIL_ADDR_MASK;
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
flags = FAULT_FLAG_ALLOW_RETRY;
if (access == VM_WRITE || (trans_exc_code & store_indication) == 0x400)
flags |= FAULT_FLAG_WRITE;
down_read(&mm->mmap_sem);
#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:
fault = VM_FAULT_BADMAP;
vma = find_vma(mm, address);
if (!vma)
goto out_up;
if (unlikely(vma->vm_start > address)) {
if (!(vma->vm_flags & VM_GROWSDOWN))
goto out_up;
if (expand_stack(vma, address))
goto out_up;
}
fault = VM_FAULT_BADACCESS;
if (unlikely(!(vma->vm_flags & access)))
goto out_up;
if (is_vm_hugetlb_page(vma))
address &= HPAGE_MASK;
fault = handle_mm_fault(mm, vma, address, flags);
if (unlikely(fault & VM_FAULT_ERROR))
goto out_up;
if (flags & FAULT_FLAG_ALLOW_RETRY) {
if (fault & VM_FAULT_MAJOR) {
tsk->maj_flt++;
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1,
regs, address);
} else {
tsk->min_flt++;
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1,
regs, address);
}
if (fault & VM_FAULT_RETRY) {
flags &= ~FAULT_FLAG_ALLOW_RETRY;
down_read(&mm->mmap_sem);
goto retry;
}
}
clear_tsk_thread_flag(tsk, TIF_PER_TRAP);
fault = 0;
out_up:
up_read(&mm->mmap_sem);
out:
return fault;
}
/*
* This routine handles page faults. It determines the address,
* and the problem, and then passes it off to one of the appropriate
* routines.
*
* error_code:
* 04 Protection -> Write-Protection (suprression)
* 10 Segment translation -> Not present (nullification)
* 11 Page translation -> Not present (nullification)
* 3b Region third trans. -> Not present (nullification)
*/
static inline void __kprobes
do_exception(struct pt_regs *regs, unsigned long error_code, int is_protection)
{
struct task_struct *tsk;
struct mm_struct *mm;
struct vm_area_struct * vma;
unsigned long address;
int user_address;
const struct exception_table_entry *fixup;
int si_code = SEGV_MAPERR;
tsk = current;
mm = tsk->mm;
if (notify_page_fault(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
SIGSEGV) == NOTIFY_STOP)
return;
/*
* 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.
*/
if (is_protection && !(S390_lowcore.trans_exc_code & 4)) {
/* Low-address protection hit in kernel mode means
NULL pointer write access in kernel mode. */
if (!(regs->psw.mask & PSW_MASK_PSTATE)) {
address = 0;
user_address = 0;
goto no_context;
}
/* Low-address protection hit in user mode 'cannot happen'. */
die ("Low-address protection", regs, error_code);
do_exit(SIGKILL);
}
/*
* get the failing address
* more specific the segment and page table portion of
* the address
*/
address = S390_lowcore.trans_exc_code & __FAIL_ADDR_MASK;
user_address = check_user_space(regs, error_code);
/*
* Verify that the fault happened in user space, that
* we are not in an interrupt and that there is a
* user context.
*/
if (user_address == 0 || in_atomic() || !mm)
goto no_context;
/*
* When we get here, the fault happened in the current
* task's user address space, so we can switch on the
* interrupts again and then search the VMAs
*/
local_irq_enable();
down_read(&mm->mmap_sem);
vma = find_vma(mm, address);
if (!vma)
goto bad_area;
if (vma->vm_start <= address)
goto good_area;
if (!(vma->vm_flags & VM_GROWSDOWN))
goto bad_area;
if (expand_stack(vma, address))
goto bad_area;
/*
* Ok, we have a good vm_area for this memory access, so
* we can handle it..
*/
good_area:
si_code = SEGV_ACCERR;
if (!is_protection) {
/* page not present, check vm flags */
if (!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE)))
goto bad_area;
} else {
if (!(vma->vm_flags & VM_WRITE))
goto bad_area;
}
survive:
if (is_vm_hugetlb_page(vma))
address &= HPAGE_MASK;
/*
* If for any reason at all we couldn't handle the fault,
* make sure we exit gracefully rather than endlessly redo
* the fault.
*/
switch (handle_mm_fault(mm, vma, address, is_protection)) {
case VM_FAULT_MINOR:
tsk->min_flt++;
break;
case VM_FAULT_MAJOR:
tsk->maj_flt++;
break;
case VM_FAULT_SIGBUS:
goto do_sigbus;
case VM_FAULT_OOM:
goto out_of_memory;
default:
BUG();
}
up_read(&mm->mmap_sem);
/*
* The instruction that caused the program check will
* be repeated. Don't signal single step via SIGTRAP.
*/
clear_tsk_thread_flag(current, TIF_SINGLE_STEP);
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:
up_read(&mm->mmap_sem);
/* User mode accesses just cause a SIGSEGV */
if (regs->psw.mask & PSW_MASK_PSTATE) {
tsk->thread.prot_addr = address;
tsk->thread.trap_no = error_code;
do_sigsegv(regs, error_code, si_code, address);
return;
}
no_context:
/* Are we prepared to handle this kernel fault? */
fixup = search_exception_tables(regs->psw.addr & __FIXUP_MASK);
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.
*/
if (user_address == 0)
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);
die("Oops", regs, error_code);
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);
if (tsk->pid == 1) {
yield();
goto survive;
}
printk("VM: killing process %s\n", tsk->comm);
if (regs->psw.mask & PSW_MASK_PSTATE)
do_exit(SIGKILL);
goto no_context;
do_sigbus:
up_read(&mm->mmap_sem);
/*
* Send a sigbus, regardless of whether we were in kernel
* or user mode.
*/
tsk->thread.prot_addr = address;
tsk->thread.trap_no = error_code;
force_sig(SIGBUS, tsk);
/* Kernel mode? Handle exceptions or die */
if (!(regs->psw.mask & PSW_MASK_PSTATE))
goto no_context;
}
/*
* This routine handles page faults. It determines the address and the
* problem, and then passes it off to one of the appropriate routines.
*
* ecr is the Exception Cause Register. Possible values are:
* 6: Protection fault (instruction access)
* 15: Protection fault (read access)
* 16: Protection fault (write access)
* 20: Page not found (instruction access)
* 24: Page not found (read access)
* 28: Page not found (write access)
*/
asmlinkage void do_page_fault(unsigned long ecr, struct pt_regs *regs)
{
struct task_struct *tsk;
struct mm_struct *mm;
struct vm_area_struct *vma;
const struct exception_table_entry *fixup;
unsigned long address;
unsigned long page;
int writeaccess;
long signr;
int code;
int fault;
if (notify_page_fault(regs, ecr))
return;
address = sysreg_read(TLBEAR);
tsk = current;
mm = tsk->mm;
signr = SIGSEGV;
code = SEGV_MAPERR;
/*
* If we're in an interrupt or have no user context, we must
* not take the fault...
*/
if (in_atomic() || !mm || regs->sr & SYSREG_BIT(GM))
goto no_context;
local_irq_enable();
down_read(&mm->mmap_sem);
vma = find_vma(mm, address);
if (!vma)
goto bad_area;
if (vma->vm_start <= address)
goto good_area;
if (!(vma->vm_flags & VM_GROWSDOWN))
goto bad_area;
if (expand_stack(vma, address))
goto bad_area;
/*
* Ok, we have a good vm_area for this memory access, so we
* can handle it...
*/
good_area:
code = SEGV_ACCERR;
writeaccess = 0;
switch (ecr) {
case ECR_PROTECTION_X:
case ECR_TLB_MISS_X:
if (!(vma->vm_flags & VM_EXEC))
goto bad_area;
break;
case ECR_PROTECTION_R:
case ECR_TLB_MISS_R:
if (!(vma->vm_flags & (VM_READ | VM_WRITE | VM_EXEC)))
goto bad_area;
break;
case ECR_PROTECTION_W:
case ECR_TLB_MISS_W:
if (!(vma->vm_flags & VM_WRITE))
goto bad_area;
writeaccess = 1;
break;
default:
panic("Unhandled case %lu in do_page_fault!", ecr);
}
/*
* If for any reason at all we couldn't handle the fault, make
* sure we exit gracefully rather than endlessly redo the
* fault.
*/
<<<<<<< HEAD
/*
* This routine handles page faults. It determines the address,
* and the problem, and then passes it off to one of the appropriate
* routines.
*
* error_code:
* 04 Protection -> Write-Protection (suprression)
* 10 Segment translation -> Not present (nullification)
* 11 Page translation -> Not present (nullification)
* 3b Region third trans. -> Not present (nullification)
*/
static inline void
do_exception(struct pt_regs *regs, unsigned long error_code, int write)
{
struct task_struct *tsk;
struct mm_struct *mm;
struct vm_area_struct *vma;
unsigned long address;
int space;
int si_code;
int fault;
if (notify_page_fault(regs, error_code))
return;
tsk = current;
mm = tsk->mm;
/* get the failing address and the affected space */
address = S390_lowcore.trans_exc_code & __FAIL_ADDR_MASK;
space = check_space(tsk);
/*
* Verify that the fault happened in user space, that
* we are not in an interrupt and that there is a
* user context.
*/
if (unlikely(space == 0 || in_atomic() || !mm))
goto no_context;
/*
* When we get here, the fault happened in the current
* task's user address space, so we can switch on the
* interrupts again and then search the VMAs
*/
local_irq_enable();
down_read(&mm->mmap_sem);
si_code = SEGV_MAPERR;
vma = find_vma(mm, address);
if (!vma)
goto bad_area;
#ifdef CONFIG_S390_EXEC_PROTECT
if (unlikely((space == 2) && !(vma->vm_flags & VM_EXEC)))
if (!signal_return(mm, regs, address, error_code))
/*
* signal_return() has done an up_read(&mm->mmap_sem)
* if it returns 0.
*/
return;
#endif
if (vma->vm_start <= address)
goto good_area;
if (!(vma->vm_flags & VM_GROWSDOWN))
goto bad_area;
if (expand_stack(vma, address))
goto bad_area;
/*
* Ok, we have a good vm_area for this memory access, so
* we can handle it..
*/
good_area:
si_code = SEGV_ACCERR;
if (!write) {
/* page not present, check vm flags */
if (!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE)))
goto bad_area;
} else {
if (!(vma->vm_flags & VM_WRITE))
goto bad_area;
}
survive:
/*
* If for any reason at all we couldn't handle the fault,
* make sure we exit gracefully rather than endlessly redo
* the fault.
*/
fault = handle_mm_fault(mm, vma, address, write);
if (unlikely(fault & VM_FAULT_ERROR)) {
if (fault & VM_FAULT_OOM) {
if (do_out_of_memory(regs, error_code, address))
goto survive;
return;
} else if (fault & VM_FAULT_SIGBUS) {
do_sigbus(regs, error_code, address);
return;
}
BUG();
}
if (fault & VM_FAULT_MAJOR)
tsk->maj_flt++;
else
tsk->min_flt++;
up_read(&mm->mmap_sem);
/*
* The instruction that caused the program check will
* be repeated. Don't signal single step via SIGTRAP.
*/
clear_tsk_thread_flag(tsk, TIF_SINGLE_STEP);
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:
up_read(&mm->mmap_sem);
/* User mode accesses just cause a SIGSEGV */
if (regs->psw.mask & PSW_MASK_PSTATE) {
tsk->thread.prot_addr = address;
tsk->thread.trap_no = error_code;
do_sigsegv(regs, error_code, si_code, address);
return;
}
no_context:
do_no_context(regs, error_code, address);
}
/*
* This routine handles page faults. It determines the address,
* and the problem, and then passes it off to one of the appropriate
* routines.
*
* interruption code (int_code):
* 04 Protection -> Write-Protection (suprression)
* 10 Segment translation -> Not present (nullification)
* 11 Page translation -> Not present (nullification)
* 3b Region third trans. -> Not present (nullification)
*/
static inline int do_exception(struct pt_regs *regs, int access)
{
struct gmap *gmap;
struct task_struct *tsk;
struct mm_struct *mm;
struct vm_area_struct *vma;
enum fault_type type;
unsigned long trans_exc_code;
unsigned long address;
unsigned int flags;
int fault;
tsk = current;
/*
* The instruction that caused the program check has
* been nullified. Don't signal single step via SIGTRAP.
*/
clear_pt_regs_flag(regs, PIF_PER_TRAP);
if (notify_page_fault(regs))
return 0;
mm = tsk->mm;
trans_exc_code = regs->int_parm_long;
/*
* Verify that the fault happened in user space, that
* we are not in an interrupt and that there is a
* user context.
*/
fault = VM_FAULT_BADCONTEXT;
type = get_fault_type(regs);
switch (type) {
case KERNEL_FAULT:
goto out;
case VDSO_FAULT:
fault = VM_FAULT_BADMAP;
goto out;
case USER_FAULT:
case GMAP_FAULT:
if (faulthandler_disabled() || !mm)
goto out;
break;
}
address = trans_exc_code & __FAIL_ADDR_MASK;
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE;
if (user_mode(regs))
flags |= FAULT_FLAG_USER;
if (access == VM_WRITE || (trans_exc_code & store_indication) == 0x400)
flags |= FAULT_FLAG_WRITE;
down_read(&mm->mmap_sem);
gmap = NULL;
if (IS_ENABLED(CONFIG_PGSTE) && type == GMAP_FAULT) {
gmap = (struct gmap *) S390_lowcore.gmap;
current->thread.gmap_addr = address;
current->thread.gmap_write_flag = !!(flags & FAULT_FLAG_WRITE);
current->thread.gmap_int_code = regs->int_code & 0xffff;
address = __gmap_translate(gmap, address);
if (address == -EFAULT) {
fault = VM_FAULT_BADMAP;
goto out_up;
}
if (gmap->pfault_enabled)
flags |= FAULT_FLAG_RETRY_NOWAIT;
}
retry:
fault = VM_FAULT_BADMAP;
vma = find_vma(mm, address);
if (!vma)
goto out_up;
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;
if (unlikely(!(vma->vm_flags & access)))
goto out_up;
if (is_vm_hugetlb_page(vma))
address &= HPAGE_MASK;
/*
* If for any reason at all we couldn't handle the fault,
* make sure we exit gracefully rather than endlessly redo
* the fault.
*/
fault = handle_mm_fault(vma, address, flags);
/* No reason to continue if interrupted by SIGKILL. */
if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) {
fault = VM_FAULT_SIGNAL;
goto out;
}
if (unlikely(fault & VM_FAULT_ERROR))
goto out_up;
/*
* 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++;
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1,
regs, address);
} else {
tsk->min_flt++;
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1,
regs, address);
}
if (fault & VM_FAULT_RETRY) {
if (IS_ENABLED(CONFIG_PGSTE) && gmap &&
(flags & FAULT_FLAG_RETRY_NOWAIT)) {
/* FAULT_FLAG_RETRY_NOWAIT has been set,
* mmap_sem has not been released */
current->thread.gmap_pfault = 1;
fault = VM_FAULT_PFAULT;
goto out_up;
}
/* Clear FAULT_FLAG_ALLOW_RETRY to avoid any risk
* of starvation. */
flags &= ~(FAULT_FLAG_ALLOW_RETRY |
FAULT_FLAG_RETRY_NOWAIT);
flags |= FAULT_FLAG_TRIED;
down_read(&mm->mmap_sem);
goto retry;
}
}
if (IS_ENABLED(CONFIG_PGSTE) && gmap) {
address = __gmap_link(gmap, current->thread.gmap_addr,
address);
if (address == -EFAULT) {
fault = VM_FAULT_BADMAP;
goto out_up;
}
if (address == -ENOMEM) {
fault = VM_FAULT_OOM;
goto out_up;
}
}
fault = 0;
out_up:
up_read(&mm->mmap_sem);
out:
return fault;
}
