The patch titled
Subject: uprobe: collapse THP pmd after removing all uprobes
has been removed from the -mm tree. Its filename was
uprobe-collapse-thp-pmd-after-removing-all-uprobes.patch
This patch was dropped because it was merged into mainline or a subsystem tree
------------------------------------------------------
From: Song Liu <songliubraving@xxxxxx>
Subject: uprobe: collapse THP pmd after removing all uprobes
After all uprobes are removed from the huge page (with PTE pgtable), it is
possible to collapse the pmd and benefit from THP again. This patch does
the collapse by calling collapse_pte_mapped_thp().
Link: http://lkml.kernel.org/r/20190815164525.1848545-7-songliubraving@xxxxxx
Signed-off-by: Song Liu <songliubraving@xxxxxx>
Acked-by: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx>
Reported-by: kbuild test robot <lkp@xxxxxxxxx>
Reviewed-by: Oleg Nesterov <oleg@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
kernel/events/uprobes.c | 9 +++++++++
1 file changed, 9 insertions(+)
--- a/kernel/events/uprobes.c~uprobe-collapse-thp-pmd-after-removing-all-uprobes
+++ a/kernel/events/uprobes.c
@@ -26,6 +26,7 @@
#include <linux/percpu-rwsem.h>
#include <linux/task_work.h>
#include <linux/shmem_fs.h>
+#include <linux/khugepaged.h>
#include <linux/uprobes.h>
@@ -472,6 +473,7 @@ int uprobe_write_opcode(struct arch_upro
struct page *old_page, *new_page;
struct vm_area_struct *vma;
int ret, is_register, ref_ctr_updated = 0;
+ bool orig_page_huge = false;
is_register = is_swbp_insn(&opcode);
uprobe = container_of(auprobe, struct uprobe, arch);
@@ -529,6 +531,9 @@ retry:
/* let go new_page */
put_page(new_page);
new_page = NULL;
+
+ if (PageCompound(orig_page))
+ orig_page_huge = true;
}
put_page(orig_page);
}
@@ -547,6 +552,10 @@ put_old:
if (ret && is_register && ref_ctr_updated)
update_ref_ctr(uprobe, mm, -1);
+ /* try collapse pmd for compound page */
+ if (!ret && orig_page_huge)
+ collapse_pte_mapped_thp(mm, vaddr);
+
return ret;
}
_
Patches currently in -mm which might be from songliubraving@xxxxxx are
