On Fri, Oct 12, 2018 at 05:42:24PM +0100, Anton Ivanov wrote:

On 10/12/18 3:48 PM, Anton Ivanov wrote:

quoted

On 12/10/2018 15:37, Kirill A. Shutemov wrote:

quoted

On Fri, Oct 12, 2018 at 03:09:49PM +0100, Anton Ivanov wrote:

quoted

On 10/12/18 2:37 AM, Joel Fernandes (Google) wrote:

quoted

Android needs to mremap large regions of memory during
memory management
related operations. The mremap system call can be really
slow if THP is
not enabled. The bottleneck is move_page_tables, which is copying each
pte at a time, and can be really slow across a large map.
Turning on THP
may not be a viable option, and is not for us. This patch
speeds up the
performance for non-THP system by copying at the PMD level
when possible.

The speed up is three orders of magnitude. On a 1GB mremap, the mremap
completion times drops from 160-250 millesconds to 380-400
microseconds.

Before:
Total mremap time for 1GB data: 242321014 nanoseconds.
Total mremap time for 1GB data: 196842467 nanoseconds.
Total mremap time for 1GB data: 167051162 nanoseconds.

After:
Total mremap time for 1GB data: 385781 nanoseconds.
Total mremap time for 1GB data: 388959 nanoseconds.
Total mremap time for 1GB data: 402813 nanoseconds.

Incase THP is enabled, the optimization is skipped. I also flush the
tlb every time we do this optimization since I couldn't find a way to
determine if the low-level PTEs are dirty. It is seen that the cost of
doing so is not much compared the improvement, on both
x86-64 and arm64.

Cc: minchan@kernel.org
Cc: pantin@google.com
Cc: hughd@google.com
Cc: lokeshgidra@google.com
Cc: dancol@google.com
Cc: mhocko@kernel.org
Cc: kirill@shutemov.name
Cc: akpm@linux-foundation.org
Signed-off-by: Joel Fernandes (Google) <redacted>
---
   mm/mremap.c | 62
+++++++++++++++++++++++++++++++++++++++++++++++++++++
   1 file changed, 62 insertions(+)

diff --git a/mm/mremap.c b/mm/mremap.c
index 9e68a02a52b1..d82c485822ef 100644
--- a/mm/mremap.c
+++ b/mm/mremap.c

@@ -191,6 +191,54 @@ static void move_ptes(struct

vm_area_struct *vma, pmd_t *old_pmd,
           drop_rmap_locks(vma);
   }
+static bool move_normal_pmd(struct vm_area_struct *vma,
unsigned long old_addr,
+          unsigned long new_addr, unsigned long old_end,
+          pmd_t *old_pmd, pmd_t *new_pmd, bool *need_flush)
+{
+    spinlock_t *old_ptl, *new_ptl;
+    struct mm_struct *mm = vma->vm_mm;
+
+    if ((old_addr & ~PMD_MASK) || (new_addr & ~PMD_MASK)
+        || old_end - old_addr < PMD_SIZE)
+        return false;
+
+    /*
+     * The destination pmd shouldn't be established, free_pgtables()
+     * should have release it.
+     */
+    if (WARN_ON(!pmd_none(*new_pmd)))
+        return false;
+
+    /*
+     * We don't have to worry about the ordering of src and dst
+     * ptlocks because exclusive mmap_sem prevents deadlock.
+     */
+    old_ptl = pmd_lock(vma->vm_mm, old_pmd);
+    if (old_ptl) {
+        pmd_t pmd;
+
+        new_ptl = pmd_lockptr(mm, new_pmd);
+        if (new_ptl != old_ptl)
+            spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING);
+
+        /* Clear the pmd */
+        pmd = *old_pmd;
+        pmd_clear(old_pmd);
+
+        VM_BUG_ON(!pmd_none(*new_pmd));
+
+        /* Set the new pmd */
+        set_pmd_at(mm, new_addr, new_pmd, pmd);

UML does not have set_pmd_at at all

Every architecture does. :)

I tried to build it patching vs 4.19-rc before I made this statement and
ran into that.

Presently it does not.

https://elixir.bootlin.com/linux/v4.19-rc7/ident/set_pmd_at - UML is not
on the list.

Once this problem as well as the omissions in the include changes for UML in
patch one have been fixed it appears to be working.

What it needs is attached.

Well, the optization is only suitable for arch that has 3 or more levels
of page tables. Otherwise it will not have [non-folded] pmd.

And in this case arch/um already should have set_pmd_at(), see
3_LEVEL_PGTABLES.

To port on 2-level paging, it has to be handled on pgd level. It
complicates the code and will not bring much value.

-- 
 Kirill A. Shutemov