For CPU with I/D separate cache, flush_cache_range/flush_tlb_range
checks vma->vm_flags if it is instruction related or not.
I think that change_protection should be called with old vma->vm_flags
so that it can flush instruction cache/TLB correctly.
Here is a patch.
Signed-off-by: NIIBE Yutaka <gniibe@xxxxxxxx>
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 4de5468..661c818 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -191,7 +191,6 @@ success:
* vm_flags and vm_page_prot are protected by the mmap_sem
* held in write mode.
*/
- vma->vm_flags = newflags;
vma->vm_page_prot = vm_get_page_prot(newflags);
if (vma_wants_writenotify(vma)) {
vma->vm_page_prot = vm_get_page_prot(newflags & ~VM_SHARED);
@@ -202,6 +201,7 @@ success:
hugetlb_change_protection(vma, start, end, vma->vm_page_prot);
else
change_protection(vma, start, end, vma->vm_page_prot, dirty_accountable);
+ vma->vm_flags = newflags;
vm_stat_account(mm, oldflags, vma->vm_file, -nrpages);
vm_stat_account(mm, newflags, vma->vm_file, nrpages);
return 0;
-
To unsubscribe from this list: send the line "unsubscribe linux-arch" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
