Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <58ba1aec-f42b-f896-617f-a3720c09b30f@samsung.com>
Date:   Fri, 14 Apr 2023 17:00:19 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
        Thunderbird/102.10.0
Subject: Re: [RFC 0/4] convert create_page_buffers to create_folio_buffers
Content-Language: en-US
To:     Hannes Reinecke <hare@suse.de>, <brauner@kernel.org>,
        <willy@infradead.org>, <viro@zeniv.linux.org.uk>,
        <akpm@linux-foundation.org>
CC:     <linux-fsdevel@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
        <mcgrof@kernel.org>, <gost.dev@samsung.com>
From:   Pankaj Raghav <p.raghav@samsung.com>
In-Reply-To: <1e68a118-d177-a218-5139-c8f13793dbbf@suse.de>
Content-Transfer-Encoding: 8bit
Content-Type: text/plain; charset="utf-8"
CMS-TYPE: 201P
References: <CGME20230414110825eucas1p1ed4d16627889ef8542dfa31b1183063d@eucas1p1.samsung.com>
        <20230414110821.21548-1-p.raghav@samsung.com>
        <1e68a118-d177-a218-5139-c8f13793dbbf@suse.de>
Precedence: bulk

>> Pankaj Raghav (4):
>>    fs/buffer: add set_bh_folio helper
>>    buffer: add alloc_folio_buffers() helper
>>    fs/buffer: add folio_create_empty_buffers helper
>>    fs/buffer: convert create_page_buffers to create_folio_buffers
>>
>>   fs/buffer.c                 | 131 +++++++++++++++++++++++++++++++++---
>>   include/linux/buffer_head.h |   4 ++
>>   2 files changed, 125 insertions(+), 10 deletions(-)
>>
> Funnily enough, I've been tinkering along the same lines, and ended up with pretty similar patches.
> I've had to use two additional patches to get my modified 'brd' driver off the ground with logical
> blocksize of 16k:

Good to know that we are working on a similar direction here.

> - mm/filemap: allocate folios according to the blocksize
>   (will be sending the patch separately)
> - Modify read_folio() to use the correct order:
> 
> @@ -2333,13 +2395,15 @@ int block_read_full_folio(struct folio *folio, get_block_t *get_block)
>         if (IS_ENABLED(CONFIG_FS_VERITY) && IS_VERITY(inode))
>                 limit = inode->i_sb->s_maxbytes;
> 
> -       VM_BUG_ON_FOLIO(folio_test_large(folio), folio);
> -
>         head = create_folio_buffers(folio, inode, 0);
>         blocksize = head->b_size;
>         bbits = block_size_bits(blocksize);
> 
> -       iblock = (sector_t)folio->index << (PAGE_SHIFT - bbits);
> +       if (WARN_ON(PAGE_SHIFT < bbits)) {
> +               iblock = (sector_t)folio->index >> (bbits - PAGE_SHIFT);
> +       } else {
> +               iblock = (sector_t)folio->index << (PAGE_SHIFT - bbits);
> +       }
>         lblock = (limit+blocksize-1) >> bbits;
>         bh = head;
>         nr = 0;
> 
> 
> With that (and my modified brd driver) I've been able to set the logical blocksize to 16k for brd
> and have it happily loaded.

I will give your patches a try as well.