[PATCH] f2fs: Delete f2fs_copy_page() and replace with memcpy_page()

[PATCH] f2fs: Delete f2fs_copy_page() and replace with memcpy_page()

[Fabio M. De Francesco]

f2fs_copy_page() is a wrapper around two kmap() + one memcpy() from/to
the mapped pages. It unnecessarily duplicates a kernel API and it makes
use of kmap(), which is being deprecated in favor of kmap_local_page().

Two main problems with kmap(): (1) It comes with an overhead as mapping
space is restricted and protected by a global lock for synchronization and
(2) it also requires global TLB invalidation when the kmap’s pool wraps
and it might block when the mapping space is fully utilized until a slot
becomes available.

With kmap_local_page() the mappings are per thread, CPU local, can take
page faults, and can be called from any context (including interrupts).
It is faster than kmap() in kernels with HIGHMEM enabled. Therefore, its
use in __clone_blkaddrs() is safe and should be preferred.

Delete f2fs_copy_page() and use a plain memcpy_page() in the only one
site calling the removed function. memcpy_page() avoids open coding two
kmap_local_page() + one memcpy() between the two kernel virtual addresses.

Suggested-by: Christoph Hellwig <hch@infradead.org>
Suggested-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>
---

This patch extends the scope and replaces "f2fs: Use memcpy_page() in
f2fs_copy_page()", as suggested by Christoph Hellwig (thanks!):
https://lore.kernel.org/lkml/YtOyWOKFN9ramUyb@infradead.org/

 fs/f2fs/f2fs.h | 10 ----------
 fs/f2fs/file.c |  2 +-
 2 files changed, 1 insertion(+), 11 deletions(-)

diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index d9bbecd008d2..52be3e23ae7c 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -2696,16 +2696,6 @@ static inline struct page *f2fs_pagecache_get_page(
 	return pagecache_get_page(mapping, index, fgp_flags, gfp_mask);
 }
 
-static inline void f2fs_copy_page(struct page *src, struct page *dst)
-{
-	char *src_kaddr = kmap(src);
-	char *dst_kaddr = kmap(dst);
-
-	memcpy(dst_kaddr, src_kaddr, PAGE_SIZE);
-	kunmap(dst);
-	kunmap(src);
-}
-
 static inline void f2fs_put_page(struct page *page, int unlock)
 {
 	if (!page)
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index bd14cef1b08f..8d984aeb270f 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -1278,7 +1278,7 @@ static int __clone_blkaddrs(struct inode *src_inode, struct inode *dst_inode,
 				f2fs_put_page(psrc, 1);
 				return PTR_ERR(pdst);
 			}
-			f2fs_copy_page(psrc, pdst);
+			memcpy_page(pdst, 0, psrc, 0, PAGE_SIZE);
 			set_page_dirty(pdst);
 			f2fs_put_page(pdst, 1);
 			f2fs_put_page(psrc, 1);
-- 
2.37.1

Re: [PATCH] f2fs: Delete f2fs_copy_page() and replace with memcpy_page()

[Christoph Hellwig]

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>

Re: [PATCH] f2fs: Delete f2fs_copy_page() and replace with memcpy_page()

[Chao Yu]

On 2022/7/17 16:36, Fabio M. De Francesco wrote:
> f2fs_copy_page() is a wrapper around two kmap() + one memcpy() from/to
> the mapped pages. It unnecessarily duplicates a kernel API and it makes
> use of kmap(), which is being deprecated in favor of kmap_local_page().
> 
> Two main problems with kmap(): (1) It comes with an overhead as mapping
> space is restricted and protected by a global lock for synchronization and
> (2) it also requires global TLB invalidation when the kmap’s pool wraps
> and it might block when the mapping space is fully utilized until a slot
> becomes available.
> 
> With kmap_local_page() the mappings are per thread, CPU local, can take
> page faults, and can be called from any context (including interrupts).
> It is faster than kmap() in kernels with HIGHMEM enabled. Therefore, its
> use in __clone_blkaddrs() is safe and should be preferred.
> 
> Delete f2fs_copy_page() and use a plain memcpy_page() in the only one
> site calling the removed function. memcpy_page() avoids open coding two
> kmap_local_page() + one memcpy() between the two kernel virtual addresses.
> 
> Suggested-by: Christoph Hellwig <hch@infradead.org>
> Suggested-by: Ira Weiny <ira.weiny@intel.com>
> Signed-off-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>

Reviewed-by: Chao Yu <chao@kernel.org>

Thanks,