[PATCH AUTOSEL 6.1 20/22] block: Allow bio_iov_iter_get_pages() with bio->bi_bdev unset

[PATCH AUTOSEL 6.1 20/22] block: Allow bio_iov_iter_get_pages() with bio->bi_bdev unset

[Sasha Levin]

From: Kent Overstreet <kent.overstreet@linux.dev>

[ Upstream commit 168145f617d57bf4e474901b7ffa869337a802e6 ]

bio_iov_iter_get_pages() trims the IO based on the block size of the
block device the IO will be issued to.

However, bcachefs is a multi device filesystem; when we're creating the
bio we don't yet know which block device the bio will be submitted to -
we have to handle the alignment checks elsewhere.

Thus this is needed to avoid a null ptr deref.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: linux-block@vger.kernel.org
Link: https://lore.kernel.org/r/20230813182636.2966159-3-kent.overstreet@linux.dev
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 block/bio.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/block/bio.c b/block/bio.c
index d5cd825d6efc0..54c8d11f09663 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -1198,7 +1198,7 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter)
 	struct page **pages = (struct page **)bv;
 	ssize_t size, left;
 	unsigned len, i = 0;
-	size_t offset, trim;
+	size_t offset;
 	int ret = 0;
 
 	/*
@@ -1223,10 +1223,12 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter)
 
 	nr_pages = DIV_ROUND_UP(offset + size, PAGE_SIZE);
 
-	trim = size & (bdev_logical_block_size(bio->bi_bdev) - 1);
-	iov_iter_revert(iter, trim);
+	if (bio->bi_bdev) {
+		size_t trim = size & (bdev_logical_block_size(bio->bi_bdev) - 1);
+		iov_iter_revert(iter, trim);
+		size -= trim;
+	}
 
-	size -= trim;
 	if (unlikely(!size)) {
 		ret = -EFAULT;
 		goto out;
-- 
2.40.1

Re: [PATCH AUTOSEL 6.1 20/22] block: Allow bio_iov_iter_get_pages() with bio->bi_bdev unset

[Pavel Machek]

[-- Attachment #1: Type: text/plain, Size: 1569 bytes --]

Hi!

> From: Kent Overstreet <kent.overstreet@linux.dev>
> 
> [ Upstream commit 168145f617d57bf4e474901b7ffa869337a802e6 ]
> 
> bio_iov_iter_get_pages() trims the IO based on the block size of the
> block device the IO will be issued to.
> 
> However, bcachefs is a multi device filesystem; when we're creating the
> bio we don't yet know which block device the bio will be submitted to -
> we have to handle the alignment checks elsewhere.
> 
> Thus this is needed to avoid a null ptr deref.

We are certainly not going to backport bcachefs to -stable. Please
drop.

BR,
								Pavel

> +++ b/block/bio.c
> @@ -1198,7 +1198,7 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter)
>  	struct page **pages = (struct page **)bv;
>  	ssize_t size, left;
>  	unsigned len, i = 0;
> -	size_t offset, trim;
> +	size_t offset;
>  	int ret = 0;
>  
>  	/*
> @@ -1223,10 +1223,12 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter)
>  
>  	nr_pages = DIV_ROUND_UP(offset + size, PAGE_SIZE);
>  
> -	trim = size & (bdev_logical_block_size(bio->bi_bdev) - 1);
> -	iov_iter_revert(iter, trim);
> +	if (bio->bi_bdev) {
> +		size_t trim = size & (bdev_logical_block_size(bio->bi_bdev) - 1);
> +		iov_iter_revert(iter, trim);
> +		size -= trim;
> +	}
>  
> -	size -= trim;
>  	if (unlikely(!size)) {
>  		ret = -EFAULT;
>  		goto out;

-- 
DENX Software Engineering GmbH,        Managing Director: Erika Unter
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

Re: [PATCH AUTOSEL 6.1 20/22] block: Allow bio_iov_iter_get_pages() with bio->bi_bdev unset

[Jens Axboe]

On 9/11/23 3:54 AM, Pavel Machek wrote:
> Hi!
> 
>> From: Kent Overstreet <kent.overstreet@linux.dev>
>>
>> [ Upstream commit 168145f617d57bf4e474901b7ffa869337a802e6 ]
>>
>> bio_iov_iter_get_pages() trims the IO based on the block size of the
>> block device the IO will be issued to.
>>
>> However, bcachefs is a multi device filesystem; when we're creating the
>> bio we don't yet know which block device the bio will be submitted to -
>> we have to handle the alignment checks elsewhere.
>>
>> Thus this is needed to avoid a null ptr deref.
> 
> We are certainly not going to backport bcachefs to -stable. Please
> drop.

I already said as much 3 days ago:

https://lore.kernel.org/all/cd341326-cfaf-4796-8894-2241e7b630d9@kernel.dk/

but didn't hear back yet.

-- 
Jens Axboe

Re: [PATCH AUTOSEL 6.1 20/22] block: Allow bio_iov_iter_get_pages() with bio->bi_bdev unset

[Sasha Levin]

On Mon, Sep 11, 2023 at 07:15:18AM -0600, Jens Axboe wrote:
>On 9/11/23 3:54 AM, Pavel Machek wrote:
>> Hi!
>>
>>> From: Kent Overstreet <kent.overstreet@linux.dev>
>>>
>>> [ Upstream commit 168145f617d57bf4e474901b7ffa869337a802e6 ]
>>>
>>> bio_iov_iter_get_pages() trims the IO based on the block size of the
>>> block device the IO will be issued to.
>>>
>>> However, bcachefs is a multi device filesystem; when we're creating the
>>> bio we don't yet know which block device the bio will be submitted to -
>>> we have to handle the alignment checks elsewhere.
>>>
>>> Thus this is needed to avoid a null ptr deref.
>>
>> We are certainly not going to backport bcachefs to -stable. Please
>> drop.
>
>I already said as much 3 days ago:
>
>https://lore.kernel.org/all/cd341326-cfaf-4796-8894-2241e7b630d9@kernel.dk/
>
>but didn't hear back yet.

Dropped, and sorry for lack of timely replies on this - I let them
collect feedback for a few weeks before going through it and dropping
what needs to be dropped.

-- 
Thanks,
Sasha