[RFC PATCH 0/4] hugetlbfs: add fallocate support

[RFC PATCH 0/4] hugetlbfs: add fallocate support

[Mike Kravetz]

hugetlbfs is used today by applications that want a high degree of
control over huge page usage.  Often, large hugetlbfs files are used
to map a large number huge pages into the application processes.
The applications know when page ranges within these large files will
no longer be used, and ideally would like to release them back to
the subpool or global pools for other uses.  The fallocate() system
call provides an interface for preallocation and hole punching within
files.  This patch set adds fallocate functionality to hugetlbfs.

Mike Kravetz (4):
  hugetlbfs: truncate_hugepages() takes a range of pages
  hugetlbfs: New huge_add_to_page_cache helper routine
  hugetlbfs: add hugetlbfs_fallocate()
  mm: madvise allow remove operation for hugetlbfs

 fs/hugetlbfs/inode.c    | 164 ++++++++++++++++++++++++++++++++++++++++++++++--
 include/linux/hugetlb.h |   5 ++
 mm/hugetlb.c            |  29 ++++++---
 mm/madvise.c            |   2 +-
 4 files changed, 185 insertions(+), 15 deletions(-)

-- 
2.1.0

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[RFC PATCH 1/4] hugetlbfs: truncate_hugepages() takes a range of pages

[Mike Kravetz]

Modify truncate_hugepages() to take a range of pages (start, end)
instead of simply start.  If the value of end is -1, this indicates
the end of the range is the end of the file.  This functionality
will be used for fallocate hole punching.

Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
---
 fs/hugetlbfs/inode.c | 25 +++++++++++++++++++++----
 1 file changed, 21 insertions(+), 4 deletions(-)

diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index c274aca..d5b67fd 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -324,11 +324,12 @@ static void truncate_huge_page(struct page *page)
 	delete_from_page_cache(page);
 }
 
-static void truncate_hugepages(struct inode *inode, loff_t lstart)
+static void truncate_hugepages(struct inode *inode, loff_t lstart, loff_t lend)
 {
 	struct hstate *h = hstate_inode(inode);
 	struct address_space *mapping = &inode->i_data;
 	const pgoff_t start = lstart >> huge_page_shift(h);
+	const pgoff_t end = lend >> huge_page_shift(h);
 	struct pagevec pvec;
 	pgoff_t next;
 	int i, freed = 0;
@@ -336,7 +337,19 @@ static void truncate_hugepages(struct inode *inode, loff_t lstart)
 	pagevec_init(&pvec, 0);
 	next = start;
 	while (1) {
-		if (!pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE)) {
+		long lookup_nr = PAGEVEC_SIZE;
+
+		/*
+		 * Make sure to never grab more pages that we
+		 * might possibly need.
+		 */
+		if (end - start < lookup_nr)
+			lookup_nr = end - start;
+		/*
+		 * This pagevec_lookup() may return pages past 'end',
+		 * so we must check for page->index > end.
+		 */
+		if (!pagevec_lookup(&pvec, mapping, next, lookup_nr)) {
 			if (next == start)
 				break;
 			next = start;
@@ -347,6 +360,10 @@ static void truncate_hugepages(struct inode *inode, loff_t lstart)
 			struct page *page = pvec.pages[i];
 
 			lock_page(page);
+			if (page->index >= end) {
+				unlock_page(page);
+				break;
+			}
 			if (page->index > next)
 				next = page->index;
 			++next;
@@ -364,7 +381,7 @@ static void hugetlbfs_evict_inode(struct inode *inode)
 {
 	struct resv_map *resv_map;
 
-	truncate_hugepages(inode, 0);
+	truncate_hugepages(inode, 0, -1);
 	resv_map = (struct resv_map *)inode->i_mapping->private_data;
 	/* root inode doesn't have the resv_map, so we should check it */
 	if (resv_map)
@@ -410,7 +427,7 @@ static int hugetlb_vmtruncate(struct inode *inode, loff_t offset)
 	if (!RB_EMPTY_ROOT(&mapping->i_mmap))
 		hugetlb_vmtruncate_list(&mapping->i_mmap, pgoff);
 	i_mmap_unlock_write(mapping);
-	truncate_hugepages(inode, offset);
+	truncate_hugepages(inode, offset, -1);
 	return 0;
 }
 
-- 
2.1.0

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[RFC PATCH 2/4] hugetlbfs: New huge_add_to_page_cache helper routine

[Mike Kravetz]

Currently, there is  only a single place where hugetlbfs pages are
added to the page cache.  The new fallocate code be adding a second
one, so break the functionality out into its own helper.

Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
---
 include/linux/hugetlb.h |  2 ++
 mm/hugetlb.c            | 27 ++++++++++++++++++---------
 2 files changed, 20 insertions(+), 9 deletions(-)

diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 7b57850..6425945 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -325,6 +325,8 @@ struct huge_bootmem_page {
 struct page *alloc_huge_page_node(struct hstate *h, int nid);
 struct page *alloc_huge_page_noerr(struct vm_area_struct *vma,
 				unsigned long addr, int avoid_reserve);
+int huge_add_to_page_cache(struct page *page, struct address_space *mapping,
+			pgoff_t idx);
 
 /* arch callback */
 int __init alloc_bootmem_huge_page(struct hstate *h);
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index c41b2a0..7cda328 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2959,6 +2959,23 @@ static bool hugetlbfs_pagecache_present(struct hstate *h,
 	return page != NULL;
 }
 
+int huge_add_to_page_cache(struct page *page, struct address_space *mapping,
+			   pgoff_t idx)
+{
+	struct inode *inode = mapping->host;
+	struct hstate *h = hstate_inode(inode);
+	int err = add_to_page_cache(page, mapping, idx, GFP_KERNEL);
+
+	if (err)
+		return err;
+	ClearPagePrivate(page);
+
+	spin_lock(&inode->i_lock);
+	inode->i_blocks += blocks_per_huge_page(h);
+	spin_unlock(&inode->i_lock);
+	return 0;
+}
+
 static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			   struct address_space *mapping, pgoff_t idx,
 			   unsigned long address, pte_t *ptep, unsigned int flags)
@@ -3005,21 +3022,13 @@ retry:
 		__SetPageUptodate(page);
 
 		if (vma->vm_flags & VM_MAYSHARE) {
-			int err;
-			struct inode *inode = mapping->host;
-
-			err = add_to_page_cache(page, mapping, idx, GFP_KERNEL);
+			int err = huge_add_to_page_cache(page, mapping, idx);
 			if (err) {
 				put_page(page);
 				if (err == -EEXIST)
 					goto retry;
 				goto out;
 			}
-			ClearPagePrivate(page);
-
-			spin_lock(&inode->i_lock);
-			inode->i_blocks += blocks_per_huge_page(h);
-			spin_unlock(&inode->i_lock);
 		} else {
 			lock_page(page);
 			if (unlikely(anon_vma_prepare(vma))) {
-- 
2.1.0

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[RFC PATCH 3/4] hugetlbfs: add hugetlbfs_fallocate()

[Mike Kravetz]

This is based on the shmem version, but it has diverged quite
a bit.  We have no swap to worry about, nor the new file sealing.

What this allows us to do is move physical memory in and out of
a hugetlbfs file without having it mapped.  This also gives us
the ability to support MADV_REMOVE since it is currently
implemented using fallocate().  MADV_REMOVE lets us remove data
from the middle of a hugetlbfs file, which wasn't possible before.

hugetlbfs fallocate only operates on whole huge pages.

Based-on code-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
---
 fs/hugetlbfs/inode.c    | 139 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/hugetlb.h |   3 ++
 mm/hugetlb.c            |   2 +-
 3 files changed, 143 insertions(+), 1 deletion(-)

diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index d5b67fd..6d48c8f 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -12,6 +12,7 @@
 #include <linux/thread_info.h>
 #include <asm/current.h>
 #include <linux/sched.h>		/* remove ASAP */
+#include <linux/falloc.h>
 #include <linux/fs.h>
 #include <linux/mount.h>
 #include <linux/file.h>
@@ -377,6 +378,143 @@ static void truncate_hugepages(struct inode *inode, loff_t lstart, loff_t lend)
 	hugetlb_unreserve_pages(inode, start, freed);
 }
 
+static long hugetlbfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
+{
+	struct hstate *h = hstate_inode(inode);
+	unsigned long hpage_size = huge_page_size(h);
+	loff_t hole_start, hole_end;
+
+	/*
+	 * For hole punch round up the beginning offset of the hole and
+	 * round down the end.
+	 */
+	hole_start = (offset + hpage_size - 1) & ~huge_page_mask(h);
+	hole_end = (offset + len - (hpage_size - 1)) * ~huge_page_mask(h);
+
+	if ((u64)hole_end > (u64)hole_start) {
+		struct address_space *mapping = &inode->i_data;
+
+		mutex_lock(&inode->i_mutex);
+		unmap_mapping_range(mapping, hole_start, hole_end, 0);
+		truncate_hugepages(inode, hole_start, hole_end);
+		mutex_unlock(&inode->i_mutex);
+	}
+
+	return 0;
+}
+
+static long hugetlbfs_fallocate(struct file *file, int mode, loff_t offset,
+				loff_t len)
+{
+	struct inode *inode = file_inode(file);
+	struct address_space *mapping = inode->i_mapping;
+	struct hstate *h = hstate_inode(inode);
+	struct vm_area_struct pseudo_vma;
+	unsigned long hpage_size = huge_page_size(h);
+	unsigned long hpage_shift = huge_page_shift(h);
+	pgoff_t start, index, end;
+	unsigned long addr;
+	int error;
+
+	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
+		return -EOPNOTSUPP;
+
+	if (mode & FALLOC_FL_PUNCH_HOLE)
+		return hugetlbfs_punch_hole(inode, offset, len);
+
+	/*
+	 * Default preallocate case.
+	 * For this range, start is rounded down and end is rounded up.
+	 */
+	start = offset >> hpage_shift;
+	end = (offset + len + hpage_size - 1) >> hpage_shift;
+
+	mutex_lock(&inode->i_mutex);
+
+	/* We need to check rlimit even when FALLOC_FL_KEEP_SIZE */
+	error = inode_newsize_ok(inode, offset + len);
+	if (error)
+		goto out;
+
+	/*
+	 * Initialize a pseudo vma that just contains the policy used
+	 * when allocating the huge pages.  The actual policy field
+	 * (vm_policy) is determined based on the index in the loop below.
+	 */
+	memset(&pseudo_vma, 0, sizeof(struct vm_area_struct));
+	pseudo_vma.vm_start = 0;
+	pseudo_vma.vm_flags |= (VM_HUGETLB | VM_MAYSHARE);
+	pseudo_vma.vm_file = file;
+
+	/* addr is the offset within the file (zero based) */
+	addr = start * hpage_size;
+	for (index = start; index < end; index++) {
+		/*
+		 * This is supposed to be the vaddr where the page is being
+		 * faulted in, but we have no vaddr here.
+		 */
+		struct page *page;
+		int avoid_reserve = 1;
+
+		cond_resched();
+
+		/*
+		 * fallocate(2) manpage permits EINTR; we may have been
+		 * interrupted because we are using up too much memory.
+		 */
+		if (signal_pending(current)) {
+			error = -EINTR;
+			break;
+		}
+		page = find_get_page(mapping, index);
+		if (page) {
+			put_page(page);
+			continue;
+		}
+
+		/* Get policy based on index */
+		pseudo_vma.vm_policy =
+			mpol_shared_policy_lookup(&HUGETLBFS_I(inode)->policy,
+							index);
+
+		page = alloc_huge_page(&pseudo_vma, addr, avoid_reserve);
+		mpol_cond_put(pseudo_vma.vm_policy);
+		if (IS_ERR(page)) {
+			error = PTR_ERR(page);
+			goto out;
+		}
+		clear_huge_page(page, addr, pages_per_huge_page(h));
+		__SetPageUptodate(page);
+		error = huge_add_to_page_cache(page, mapping, index);
+		if (error) {
+			put_page(page);
+			/* Keep going if we see an -EEXIST */
+			if (error != -EEXIST)
+				goto out;  /* FIXME, need to free? */
+		}
+
+		/*
+		 * page_put due to reference from alloc_huge_page()
+		 * unlock_page because locked by add_to_page_cache()
+		 */
+		put_page(page);
+		unlock_page(page);
+
+		/* Increment addr for next huge page */
+		addr += hpage_size;
+	}
+
+	if (!(mode & FALLOC_FL_KEEP_SIZE) && offset + len > inode->i_size)
+		i_size_write(inode, offset + len);
+	inode->i_ctime = CURRENT_TIME;
+	spin_lock(&inode->i_lock);
+	inode->i_private = NULL;
+	spin_unlock(&inode->i_lock);
+out:
+	mutex_unlock(&inode->i_mutex);
+	return error;
+}
+
 static void hugetlbfs_evict_inode(struct inode *inode)
 {
 	struct resv_map *resv_map;
@@ -743,6 +881,7 @@ const struct file_operations hugetlbfs_file_operations = {
 	.fsync			= noop_fsync,
 	.get_unmapped_area	= hugetlb_get_unmapped_area,
 	.llseek		= default_llseek,
+	.fallocate		= hugetlbfs_fallocate,
 };
 
 static const struct inode_operations hugetlbfs_dir_inode_operations = {
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 6425945..d96b88e 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -322,6 +322,8 @@ struct huge_bootmem_page {
 #endif
 };
 
+struct page *alloc_huge_page(struct vm_area_struct *vma,
+				unsigned long addr, int avoid_reserve);
 struct page *alloc_huge_page_node(struct hstate *h, int nid);
 struct page *alloc_huge_page_noerr(struct vm_area_struct *vma,
 				unsigned long addr, int avoid_reserve);
@@ -476,6 +478,7 @@ static inline bool hugepages_supported(void)
 
 #else	/* CONFIG_HUGETLB_PAGE */
 struct hstate {};
+#define alloc_huge_page(v, a, r) NULL
 #define alloc_huge_page_node(h, nid) NULL
 #define alloc_huge_page_noerr(v, a, r) NULL
 #define alloc_bootmem_huge_page(h) NULL
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 7cda328..e130c6d 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1363,7 +1363,7 @@ static void vma_commit_reservation(struct hstate *h,
 	region_add(resv, idx, idx + 1);
 }
 
-static struct page *alloc_huge_page(struct vm_area_struct *vma,
+struct page *alloc_huge_page(struct vm_area_struct *vma,
 				    unsigned long addr, int avoid_reserve)
 {
 	struct hugepage_subpool *spool = subpool_vma(vma);
-- 
2.1.0

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[RFC PATCH 4/4] mm: madvise allow remove operation for hugetlbfs

[Mike Kravetz]

Now that we have hole punching support for hugetlbfs, we can
also support the MADV_REMOVE interface to it.

Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
---
 mm/madvise.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/madvise.c b/mm/madvise.c
index d551475..c4a1027 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -299,7 +299,7 @@ static long madvise_remove(struct vm_area_struct *vma,
 
 	*prev = NULL;	/* tell sys_madvise we drop mmap_sem */
 
-	if (vma->vm_flags & (VM_LOCKED | VM_HUGETLB))
+	if (vma->vm_flags & VM_LOCKED)
 		return -EINVAL;
 
 	f = vma->vm_file;
-- 
2.1.0

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [RFC PATCH 4/4] mm: madvise allow remove operation for hugetlbfs

[Christoph Hellwig]

On Thu, Apr 16, 2015 at 04:02:58PM -0700, Mike Kravetz wrote:
> Now that we have hole punching support for hugetlbfs, we can
> also support the MADV_REMOVE interface to it.

Meh.  Just use fallocate for any new code..

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [RFC PATCH 4/4] mm: madvise allow remove operation for hugetlbfs

[Mike Kravetz]

On 04/16/2015 11:44 PM, Christoph Hellwig wrote:
> On Thu, Apr 16, 2015 at 04:02:58PM -0700, Mike Kravetz wrote:
>> Now that we have hole punching support for hugetlbfs, we can
>> also support the MADV_REMOVE interface to it.
>
> Meh.  Just use fallocate for any new code..
>

I don't have the complete context for your comment.  Is there
a general consensus or effort underway to deprecate the use
of MADV_REMOVE?

-- 
Mike Kravetz

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>