The quilt patch titled
Subject: buffer: fix more functions for block size > PAGE_SIZE
has been removed from the -mm tree. Its filename was
buffer-fix-more-functions-for-block-size-page_size.patch
This patch was dropped because it was merged into the mm-stable branch
of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
------------------------------------------------------
From: "Matthew Wilcox (Oracle)" <willy@xxxxxxxxxxxxx>
Subject: buffer: fix more functions for block size > PAGE_SIZE
Date: Thu, 9 Nov 2023 21:06:08 +0000
Both __block_write_full_folio() and block_read_full_folio() assumed that
block size <= PAGE_SIZE. Replace the shift with a divide, which is
probably cheaper than first calculating the shift. That lets us remove
block_size_bits() as these were the last callers.
Link: https://lkml.kernel.org/r/20231109210608.2252323-8-willy@xxxxxxxxxxxxx
Signed-off-by: Matthew Wilcox (Oracle) <willy@xxxxxxxxxxxxx>
Cc: Hannes Reinecke <hare@xxxxxxx>
Cc: Luis Chamberlain <mcgrof@xxxxxxxxxx>
Cc: Pankaj Raghav <p.raghav@xxxxxxxxxxx>
Cc: Ryusuke Konishi <konishi.ryusuke@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
fs/buffer.c | 27 ++++++---------------------
1 file changed, 6 insertions(+), 21 deletions(-)
--- a/fs/buffer.c~buffer-fix-more-functions-for-block-size-page_size
+++ a/fs/buffer.c
@@ -1742,19 +1742,6 @@ unlock_page:
}
EXPORT_SYMBOL(clean_bdev_aliases);
-/*
- * Size is a power-of-two in the range 512..PAGE_SIZE,
- * and the case we care about most is PAGE_SIZE.
- *
- * So this *could* possibly be written with those
- * constraints in mind (relevant mostly if some
- * architecture has a slow bit-scan instruction)
- */
-static inline int block_size_bits(unsigned int blocksize)
-{
- return ilog2(blocksize);
-}
-
static struct buffer_head *folio_create_buffers(struct folio *folio,
struct inode *inode,
unsigned int b_state)
@@ -1807,7 +1794,7 @@ int __block_write_full_folio(struct inod
sector_t block;
sector_t last_block;
struct buffer_head *bh, *head;
- unsigned int blocksize, bbits;
+ size_t blocksize;
int nr_underway = 0;
blk_opf_t write_flags = wbc_to_write_flags(wbc);
@@ -1826,10 +1813,9 @@ int __block_write_full_folio(struct inod
bh = head;
blocksize = bh->b_size;
- bbits = block_size_bits(blocksize);
- block = (sector_t)folio->index << (PAGE_SHIFT - bbits);
- last_block = (i_size_read(inode) - 1) >> bbits;
+ block = div_u64(folio_pos(folio), blocksize);
+ last_block = div_u64(i_size_read(inode) - 1, blocksize);
/*
* Get all the dirty buffers mapped to disk addresses and
@@ -2355,7 +2341,7 @@ int block_read_full_folio(struct folio *
struct inode *inode = folio->mapping->host;
sector_t iblock, lblock;
struct buffer_head *bh, *head, *arr[MAX_BUF_PER_PAGE];
- unsigned int blocksize, bbits;
+ size_t blocksize;
int nr, i;
int fully_mapped = 1;
bool page_error = false;
@@ -2369,10 +2355,9 @@ int block_read_full_folio(struct folio *
head = folio_create_buffers(folio, inode, 0);
blocksize = head->b_size;
- bbits = block_size_bits(blocksize);
- iblock = (sector_t)folio->index << (PAGE_SHIFT - bbits);
- lblock = (limit+blocksize-1) >> bbits;
+ iblock = div_u64(folio_pos(folio), blocksize);
+ lblock = div_u64(limit + blocksize - 1, blocksize);
bh = head;
nr = 0;
i = 0;
_
Patches currently in -mm which might be from willy@xxxxxxxxxxxxx are
mm-remove-inc-dec-lruvec-page-state-functions.patch
slub-use-alloc_pages_node-in-alloc_slab_page.patch
slub-use-folio-apis-in-free_large_kmalloc.patch
slub-use-a-folio-in-__kmalloc_large_node.patch
mm-khugepaged-use-a-folio-more-in-collapse_file.patch
mm-memcontrol-remove-__mod_lruvec_page_state.patch
