mm/m68k: use general page fault accounting
authorPeter Xu <peterx@redhat.com>
Wed, 12 Aug 2020 01:38:09 +0000 (18:38 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Wed, 12 Aug 2020 17:58:03 +0000 (10:58 -0700)
Use the general page fault accounting by passing regs into
handle_mm_fault().  It naturally solve the issue of multiple page fault
accounting when page fault retry happened.

Add the missing PERF_COUNT_SW_PAGE_FAULTS perf events too.  Note, the
other two perf events (PERF_COUNT_SW_PAGE_FAULTS_[MAJ|MIN]) were done in
handle_mm_fault().

Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Link: http://lkml.kernel.org/r/20200707225021.200906-10-peterx@redhat.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
arch/m68k/mm/fault.c

index 08b35a318ebeb3c305c22290195fa79614c9ebfd..795f483b105091a6c65942cda524a54375bd2d9a 100644 (file)
@@ -12,6 +12,7 @@
 #include <linux/interrupt.h>
 #include <linux/module.h>
 #include <linux/uaccess.h>
+#include <linux/perf_event.h>
 
 #include <asm/setup.h>
 #include <asm/traps.h>
@@ -84,6 +85,8 @@ int do_page_fault(struct pt_regs *regs, unsigned long address,
 
        if (user_mode(regs))
                flags |= FAULT_FLAG_USER;
+
+       perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
 retry:
        mmap_read_lock(mm);
 
@@ -134,7 +137,7 @@ good_area:
         * the fault.
         */
 
-       fault = handle_mm_fault(vma, address, flags, NULL);
+       fault = handle_mm_fault(vma, address, flags, regs);
        pr_debug("handle_mm_fault returns %x\n", fault);
 
        if (fault_signal_pending(fault, regs))
@@ -150,16 +153,7 @@ good_area:
                BUG();
        }
 
-       /*
-        * 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)
-                       current->maj_flt++;
-               else
-                       current->min_flt++;
                if (fault & VM_FAULT_RETRY) {
                        flags |= FAULT_FLAG_TRIED;