convert remaining architectures to pagefault_out_of_memory()

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Richard Henderson <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Richard Henderson <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Richard Henderson <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

---
Index: linux-2.6/arch/alpha/mm/fault.c
===================================================================

arch/alpha/mm/fault.c +0 -0

@@ -188,16 +188,10 @@ do_page_fault(unsigned long address, uns
        /* 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:
-       if (is_global_init(current)) {
-               yield();
-               down_read(&mm->mmap_sem);
-               goto survive;
-       }
-       printk(KERN_ALERT "VM: killing process %s(%d)\n",
-              current->comm, task_pid_nr(current));
       if (!user_mode(regs))
               goto no_context;
-       do_group_exit(SIGKILL);
+       pagefault_out_of_memory();
+       return;

 do_sigbus:
       /* Send a sigbus, regardless of whether we were in kernel

--
To unsubscribe from this list: send the line "unsubscribe linux-alpha" in the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Quoted 30 lines · On Thu, Apr 22, 2010 at 12:06 PM,  <[email protected]> wrote:

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Richard Henderson <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

---
Index: linux-2.6/arch/alpha/mm/fault.c
===================================================================

arch/alpha/mm/fault.c +0 -0

@@ -188,16 +188,10 @@ do_page_fault(unsigned long address, uns
        /* 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:
-       if (is_global_init(current)) {
-               yield();
-               down_read(&mm->mmap_sem);
-               goto survive;
-       }
-       printk(KERN_ALERT "VM: killing process %s(%d)\n",
-              current->comm, task_pid_nr(current));
       if (!user_mode(regs))
               goto no_context;
-       do_group_exit(SIGKILL);
+       pagefault_out_of_memory();
+       return;

Quoted 9 lines

 do_sigbus:
       /* Send a sigbus, regardless of whether we were in kernel

--
To unsubscribe from this list: send the line "unsubscribe linux-alpha" in the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Thanks, added to my tree.

Matt

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc Haavard Skinnemoen <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

---
Index: linux-2.6/arch/avr32/mm/fault.c
===================================================================

arch/avr32/mm/fault.c +0 -2

@@ -211,15 +211,10 @@ no_context:
 	 */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (is_global_init(current)) {
-		yield();

arch/avr32/mm/fault.c +1 -0

+	return;
 
 do_sigbus:
 	up_read(&mm->mmap_sem);

Quoted 23 lines · On Fri, 23 Apr 2010, [email protected] wrote:

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc Haavard Skinnemoen <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

---
Index: linux-2.6/arch/avr32/mm/fault.c
===================================================================

arch/avr32/mm/fault.c +0 -7

@@ -211,15 +211,10 @@ no_context:
 	 */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (is_global_init(current)) {
-		yield();
-		down_read(&mm->mmap_sem);
-		goto survive;
-	}
-	printk("VM: Killing process %s\n", tsk->comm);
-	if (user_mode(regs))

arch/avr32/mm/fault.c no_context: · line 216

Quoted 5 lines

arch/avr32/mm/fault.c +3 -2

-		do_group_exit(SIGKILL);
-	goto no_context;
+	pagefault_out_of_memory();
+	if (!user_mode(regs))
+		goto no_context;

We want to test for !user_mode(regs) before calling pagefault_out_of_memory().

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Mikael Starvik <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc David Howells <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc David Howells <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Tony Luck <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

---
Index: linux-2.6/arch/ia64/mm/fault.c =================================> --- linux-2.6.orig/arch/ia64/mm/fault.c +++ linux-2.6/arch/ia64/mm/fault.c

(quoted diff) +3 -9

@@ -276,13 +276,7 @@ ia64_do_page_fault (unsigned long addres
 
   out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (is_global_init(current)) {
-		yield();
-		down_read(&mm->mmap_sem);
-		goto survive;
-	}
-	printk(KERN_CRIT "VM: killing process %s\n", current->comm);
-	if (user_mode(regs))
-		do_group_exit(SIGKILL);
-	goto no_context;
+	if (!user_mode(regs))
+		goto no_context;
+	pagefault_out_of_memory();
 }

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Hirokazu Takata <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

---
Index: linux-2.6/arch/m32r/mm/fault.c
===================================================================

arch/m32r/mm/fault.c +3 -2

-		do_group_exit(SIGKILL);
-	goto no_context;
+		goto no_context;
+	pagefault_out_of_memory();
+	return;
 
 do_sigbus:
 	up_read(&mm->mmap_sem);

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Geert Uytterhoeven <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Geert Uytterhoeven <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

---
Index: linux-2.6/arch/m68k/mm/fault.c
===================================================================

arch/m68k/mm/fault.c +0 -0

@@ -180,15 +180,10 @@ good_area:
  */

 out_of_memory:
       up_read(&mm->mmap_sem);
-       if (is_global_init(current)) {
-               yield();
-               down_read(&mm->mmap_sem);
-               goto survive;
-       }
-
-       printk("VM: killing process %s\n", current->comm);
-       if (user_mode(regs))
-               do_group_exit(SIGKILL);
+       if (!user_mode(regs))
+               goto no_context;
+       pagefault_out_of_memory();
+       return;

 no_context:
       current->thread.signo = SIGBUS;

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Michal Simek <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

Quoted 4 lines · On Fri, 23 Apr 2010, [email protected] wrote:

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Quoted 4 lines

Cc [email protected]

Cc Michal Simek <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

Acked-by David Rientjes <[email protected]>

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc David Howells <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Kyle McMartin <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

Quoted 4 lines · On Fri, 23 Apr 2010, [email protected] wrote:

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Quoted 4 lines

Cc [email protected]

Cc Kyle McMartin <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

Acked-by David Rientjes <[email protected]>

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Benjamin Herrenschmidt <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

---
Index: linux-2.6/arch/powerpc/mm/fault.c
===================================================================

arch/powerpc/mm/fault.c +0 -3

@@ -359,15 +359,10 @@ bad_area_nosemaphore:
  */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (is_global_init(current)) {
-		yield();
-		down_read(&mm->mmap_sem);

Quoted 20 lines · On Fri, 23 Apr 2010, [email protected] wrote:

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Benjamin Herrenschmidt <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

---
Index: linux-2.6/arch/powerpc/mm/fault.c
===================================================================

arch/powerpc/mm/fault.c +0 -3

@@ -359,15 +359,10 @@ bad_area_nosemaphore:
  */
 out_of_memory:
 	up_read(&mm->mmap_sem);
-	if (is_global_init(current)) {
-		yield();
-		down_read(&mm->mmap_sem);

arch/powerpc/mm/fault.c bad_area_nosemaphore: · line 365

Quoted 10 lines

arch/powerpc/mm/fault.c +4 -6

-		goto survive;
-	}
-	printk("VM: killing process %s\n", current->comm);
-	if (user_mode(regs))
-		do_group_exit(SIGKILL);
-	return SIGKILL;
+	if (!user_mode(regs))
+		return SIGKILL;
+	pagefault_out_of_memory();
+	return 0;

Do we really want to return 0 and indicate that the fault was handled? It seems more consistent to do

	if (user_mode(regs))
		pagefault_out_of_memory();
	return SIGKILL;

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc Chen Liqin <[email protected]>

Cc Lennox Wu <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

As explained in commit 1c0fe6e3bd, we want to call the architecture

oom killer when getting an unexplained OOM from handle_mm_fault, rather

simply killing current.

Cc Chen Liqin <[email protected]>

Cc Lennox Wu <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

---
Index: linux-2.6/arch/score/mm/fault.c
===================================================================

arch/score/mm/fault.c +0 -6

@@ -167,15 +167,10 @@ no_context:
    */
 out_of_memory:
    up_read(&mm->mmap_sem);
-   if (is_global_init(tsk)) {
-      yield();
-      down_read(&mm->mmap_sem);
-      goto survive;
-   }
-   printk("VM: killing process %s\n", tsk->comm);

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Paul Mundt <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc [email protected]

Cc Paul Mundt <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>

As explained in commit 1c0fe6e3bd, we want to call the architecture independent oom killer when getting an unexplained OOM from handle_mm_fault, rather than simply killing current.

Cc Chris Zankel <[email protected]>

Cc [email protected]

Signed-off-by Nick Piggin <[email protected]>