Re: [PATCH 0/5] enable bs > ps for block devices

["Pankaj Raghav (Samsung)" <kernel@pankajraghav.com>]

On Fri, May 10, 2024 at 12:29:01PM +0200, hare@kernel.org wrote:
> From: Hannes Reinecke <hare@kernel.org>
> 
> Hi all,
> 
> based on the patch series from Pankaj '[PATCH v5 00/11] enable bs > ps in XFS'
> it's now quite simple to enable support for block devices with block sizes
> larger than page size even without having to disable CONFIG_BUFFER_HEAD.
> The patchset really is just two rather trivial patches to fs/mpage,
> and two patches to remove hardcoded restrictions on the block size.
> 
> As usual, comments and reviews are welcome.
> 
> Hannes Reinecke (4):
>   fs/mpage: use blocks_per_folio instead of blocks_per_page
>   fs/mpage: avoid negative shift for large blocksize
>   block/bdev: lift restrictions on supported blocksize
>   block/bdev: enable large folio support for large logical block sizes
> 
> Pankaj Raghav (1):
>   nvme: enable logical block size > PAGE_SIZE
> 
>  block/bdev.c             | 10 ++++++---
>  drivers/nvme/host/core.c |  7 +++----
>  fs/mpage.c               | 44 ++++++++++++++++++++--------------------

What about fs/buffer.c? Do we need some changes there?

One obvious place I see is:

diff --git a/fs/buffer.c b/fs/buffer.c
index 8c19e705b9c3..ae248a10a467 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -1121,7 +1121,7 @@ __getblk_slow(struct block_device *bdev, sector_t block,
 {
        /* Size must be multiple of hard sectorsize */
        if (unlikely(size & (bdev_logical_block_size(bdev)-1) ||
-                       (size < 512 || size > PAGE_SIZE))) {
+                       (size < 512))) {
                printk(KERN_ERR "getblk(): invalid block size %d requested\n",
                                        size);
                printk(KERN_ERR "logical block size: %d\n",