[PATCH] mm/gup: restore the ability to pin more than 2GB at a time

[John Hubbard <jhubbard@nvidia.com>]

commit 53ba78de064b ("mm/gup: introduce
check_and_migrate_movable_folios()") created a new constraint on the
pin_user_pages*() API family: a potentially large allocation must now
occur, internally.

A user-visible consequence has now appeared: user space can no longer
pin more than 2GB of memory anymore on x86_64. That's because, on a 4KB
PAGE_SIZE system, when user space tries to (indirectly, via a device
driver that calls pin_user_pages()) pin 2GB, this requires an allocation
of a folio pointers array of MAX_PAGE_ORDER size, which is the limit for
kmalloc().

Fix this (restore the original behavior), by using replacing
kmalloc_array() with kvmalloc_array(), which falls back to vmalloc() for
larger allocations.

Fixes: 53ba78de064b ("mm/gup: introduce check_and_migrate_movable_folios()")
Cc: linux-stable@vger.kernel.org

Cc: Vivek Kasireddy <vivek.kasireddy@intel.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Dave Airlie <airlied@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Dongwon Kim <dongwon.kim@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Junxiao Chang <junxiao.chang@intel.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Oscar Salvador <osalvador@suse.de>
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
---

This applies to mm-hotfixes-unstable (only), because it relies on my
earlier patch to this exact same location: commit 255231c75dcd mm/gup:
stop leaking pinned pages in low memory conditions.

thanks,
John Hubbard

 mm/gup.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/mm/gup.c b/mm/gup.c
index 4637dab7b54f..346186788a49 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -21,6 +21,7 @@
 #include <linux/pagevec.h>
 #include <linux/sched/mm.h>
 #include <linux/shmem_fs.h>
+#include <linux/vmalloc.h>
 
 #include <asm/mmu_context.h>
 #include <asm/tlbflush.h>
@@ -2439,7 +2440,7 @@ static long check_and_migrate_movable_pages(unsigned long nr_pages,
 	struct folio **folios;
 	long i, ret;
 
-	folios = kmalloc_array(nr_pages, sizeof(*folios), GFP_KERNEL);
+	folios = kvmalloc_array(nr_pages, sizeof(*folios), GFP_KERNEL);
 	if (!folios) {
 		unpin_user_pages(pages, nr_pages);
 		return -ENOMEM;
@@ -2450,7 +2451,7 @@ static long check_and_migrate_movable_pages(unsigned long nr_pages,
 
 	ret = check_and_migrate_movable_folios(nr_pages, folios);
 
-	kfree(folios);
+	kvfree(folios);
 	return ret;
 }
 #else

base-commit: b70a32bbebeae216a3e846e01965880b309ca173
-- 
2.47.0