public inbox for linux-btrfs@vger.kernel.org
 help / color / mirror / Atom feed
From: "Miquel Sabaté Solà" <mssola@mssola.com>
To: Qu Wenruo <wqu@suse.com>
Cc: Qu Wenruo <quwenruo.btrfs@gmx.com>,
	 dsterba@suse.com,  clm@fb.com, naohiro.aota@wdc.com,
	 linux-btrfs@vger.kernel.org, linux-kernel@vger.kernel.org,
	 kees@kernel.org
Subject: Re: [PATCH] btrfs: replace kcalloc() calls to kzalloc_objs()
Date: Tue, 24 Feb 2026 09:59:04 +0100	[thread overview]
Message-ID: <87qzqa5xgn.fsf@> (raw)
In-Reply-To: <e21ea0cf-b392-487f-843f-962efedcf10c@suse.com> (Qu Wenruo's message of "Tue, 24 Feb 2026 17:18:31 +1030")

[-- Attachment #1: Type: text/plain, Size: 7164 bytes --]

Qu Wenruo @ 2026-02-24 17:18 +1030:

> 在 2026/2/24 17:06, Miquel Sabaté Solà 写道:
>> Qu Wenruo @ 2026-02-24 15:07 +1030:
>>
>>> 在 2026/2/24 10:14, Miquel Sabaté Solà 写道:
>>>> Commit 2932ba8d9c99 ("slab: Introduce kmalloc_obj() and family")
>>>> introduced, among many others, the kzalloc_objs() helper, which has some
>>>> benefits over kcalloc().
>>>> Cc: Kees Cook <kees@kernel.org>
>>>> Signed-off-by: Miquel Sabaté Solà <mssola@mssola.com>
>>>> ---
>>>>    fs/btrfs/block-group.c       | 2 +-
>>>>    fs/btrfs/raid56.c            | 8 ++++----
>>>>    fs/btrfs/tests/zoned-tests.c | 2 +-
>>>>    fs/btrfs/volumes.c           | 6 ++----
>>>>    fs/btrfs/zoned.c             | 5 ++---
>>>>    5 files changed, 10 insertions(+), 13 deletions(-)
>>>> diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
>>>> index 37bea850b3f0..8d85b4707690 100644
>>>> --- a/fs/btrfs/block-group.c
>>>> +++ b/fs/btrfs/block-group.c
>>>> @@ -2239,7 +2239,7 @@ int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start,
>>>>    	if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK)
>>>>    		io_stripe_size = btrfs_stripe_nr_to_offset(nr_data_stripes(map));
>>>>    -	buf = kcalloc(map->num_stripes, sizeof(u64), GFP_NOFS);
>>>> +	buf = kzalloc_objs(*buf, map->num_stripes, GFP_NOFS);
>>>
>>> Not sure if we should use *buf for the type.
>>>
>>> I still remember we had some bugs related to incorrect type usage.
>> Considering the type of 'buf' and how kzalloc_objs() will resolve the
>> first argument '*buf', it should really just be equivalent to what was
>> written before.
>
> Good luck if you miss the '*' for a structure pointer, and compiler won't give
> you any warning.
>
> I still remembered that Johannes exposed such bug for me.
> Unfortunately I didn't have exact lore link for it.

That's a good point. Fortunately for us, this is what I get if I remove
the star with GCC 15.2:

fs/btrfs/block-group.c:2242:13: error: assignment to 'u64 *' {aka 'long long unsigned int *'} from incompatible pointer type 'u64 **' {aka 'long long unsigned int **'} [-Wincompatible-pointer-types]
 2242 |         buf = kmalloc_objs(buf, map->num_stripes, GFP_NOFS);
      |

Hence, it looks like these kinds of errors will be catched by the
compiler with these new helpers. This doesn't mean that we are 100% safe
from corner cases, but at least there are some guarantees.

>
>>
>>>
>>> Another thing is, we may not want to use the kzalloc version.
>>> We don't want to waste CPU time just to zero out the content meanwhile we're
>>> ensured to re-assign the contents.
>>>
>>> Thus kmalloc_objs() maybe better.
>> Yes, having a second look at this function, it looks like kmalloc_objs()
>> might just be enough. If you don't mind, I will add another commit in v2
>> translating kzalloc_objs() into kmalloc_objs() wherever I see we can do
>> this from the ones I've touched. This way we can easily revert in case
>> things go south :)
>>
>>>
>>> Thanks,
>>> Qu
>> Thanks,
>> Miquel
>>
>>>
>>>
>>>>    	if (!buf) {
>>>>    		ret = -ENOMEM;
>>>>    		goto out;
>>>> diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c
>>>> index 02105d68accb..1ebfed8f0a0a 100644
>>>> --- a/fs/btrfs/raid56.c
>>>> +++ b/fs/btrfs/raid56.c
>>>> @@ -2110,8 +2110,8 @@ static int recover_sectors(struct btrfs_raid_bio *rbio)
>>>>    	 * @unmap_array stores copy of pointers that does not get reordered
>>>>    	 * during reconstruction so that kunmap_local works.
>>>>    	 */
>>>> -	pointers = kcalloc(rbio->real_stripes, sizeof(void *), GFP_NOFS);
>>>> -	unmap_array = kcalloc(rbio->real_stripes, sizeof(void *), GFP_NOFS);
>>>> +	pointers = kzalloc_objs(*pointers, rbio->real_stripes, GFP_NOFS);
>>>> +	unmap_array = kzalloc_objs(*unmap_array, rbio->real_stripes, GFP_NOFS);
>>>>    	if (!pointers || !unmap_array) {
>>>>    		ret = -ENOMEM;
>>>>    		goto out;
>>>> @@ -2844,8 +2844,8 @@ static int recover_scrub_rbio(struct btrfs_raid_bio *rbio)
>>>>    	 * @unmap_array stores copy of pointers that does not get reordered
>>>>    	 * during reconstruction so that kunmap_local works.
>>>>    	 */
>>>> -	pointers = kcalloc(rbio->real_stripes, sizeof(void *), GFP_NOFS);
>>>> -	unmap_array = kcalloc(rbio->real_stripes, sizeof(void *), GFP_NOFS);
>>>> +	pointers = kzalloc_objs(*pointers, rbio->real_stripes, GFP_NOFS);
>>>> +	unmap_array = kzalloc_objs(*unmap_array, rbio->real_stripes, GFP_NOFS);
>>>>    	if (!pointers || !unmap_array) {
>>>>    		ret = -ENOMEM;
>>>>    		goto out;
>>>> diff --git a/fs/btrfs/tests/zoned-tests.c b/fs/btrfs/tests/zoned-tests.c
>>>> index da21c7aea31a..2bc3b14baa41 100644
>>>> --- a/fs/btrfs/tests/zoned-tests.c
>>>> +++ b/fs/btrfs/tests/zoned-tests.c
>>>> @@ -58,7 +58,7 @@ static int test_load_zone_info(struct btrfs_fs_info *fs_info,
>>>>    		return -ENOMEM;
>>>>    	}
>>>>    -	zone_info = kcalloc(test->num_stripes, sizeof(*zone_info), GFP_KERNEL);
>>>> +	zone_info = kzalloc_objs(*zone_info, test->num_stripes, GFP_KERNEL);
>>>>    	if (!zone_info) {
>>>>    		test_err("cannot allocate zone info");
>>>>    		return -ENOMEM;
>>>> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
>>>> index e15e138c515b..c0cf8f7c5a8e 100644
>>>> --- a/fs/btrfs/volumes.c
>>>> +++ b/fs/btrfs/volumes.c
>>>> @@ -5499,8 +5499,7 @@ static int calc_one_profile_avail(struct btrfs_fs_info *fs_info,
>>>>    		goto out;
>>>>    	}
>>>>    -	devices_info = kcalloc(fs_devices->rw_devices, sizeof(*devices_info),
>>>> -			       GFP_NOFS);
>>>> +	devices_info = kzalloc_objs(*devices_info, fs_devices->rw_devices, GFP_NOFS);
>>>>    	if (!devices_info) {
>>>>    		ret = -ENOMEM;
>>>>    		goto out;
>>>> @@ -6067,8 +6066,7 @@ struct btrfs_block_group *btrfs_create_chunk(struct btrfs_trans_handle *trans,
>>>>    	ctl.space_info = space_info;
>>>>    	init_alloc_chunk_ctl(fs_devices, &ctl);
>>>>    -	devices_info = kcalloc(fs_devices->rw_devices, sizeof(*devices_info),
>>>> -			       GFP_NOFS);
>>>> +	devices_info = kzalloc_objs(*devices_info, fs_devices->rw_devices, GFP_NOFS);
>>>>    	if (!devices_info)
>>>>    		return ERR_PTR(-ENOMEM);
>>>>    diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c
>>>> index ab330ec957bc..851b0de7bed7 100644
>>>> --- a/fs/btrfs/zoned.c
>>>> +++ b/fs/btrfs/zoned.c
>>>> @@ -1697,8 +1697,7 @@ static int btrfs_load_block_group_raid10(struct btrfs_block_group *bg,
>>>>    		return -EINVAL;
>>>>    	}
>>>>    -	raid0_allocs = kcalloc(map->num_stripes / map->sub_stripes,
>>>> sizeof(*raid0_allocs),
>>>> -			       GFP_NOFS);
>>>> +	raid0_allocs = kzalloc_objs(*raid0_allocs, map->num_stripes / map->sub_stripes, GFP_NOFS);
>>>>    	if (!raid0_allocs)
>>>>    		return -ENOMEM;
>>>>    @@ -1916,7 +1915,7 @@ int btrfs_load_block_group_zone_info(struct
>>>> btrfs_block_group *cache, bool new)
>>>>      	cache->physical_map = map;
>>>>    -	zone_info = kcalloc(map->num_stripes, sizeof(*zone_info), GFP_NOFS);
>>>> +	zone_info = kzalloc_objs(*zone_info, map->num_stripes, GFP_NOFS);
>>>>    	if (!zone_info) {
>>>>    		ret = -ENOMEM;
>>>>    		goto out;

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 897 bytes --]

  reply	other threads:[~2026-02-24  8:59 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-23 23:44 [PATCH] btrfs: replace kcalloc() calls to kzalloc_objs() Miquel Sabaté Solà
2026-02-24  0:06 ` Kees Cook
2026-02-24  6:23   ` Miquel Sabaté Solà
     [not found]   ` <699d43e6.170a0220.3a6e96.a235SMTPIN_ADDED_BROKEN@mx.google.com>
2026-02-24 11:29     ` David Sterba
2026-02-24 12:23       ` Miquel Sabaté Solà
2026-02-24  4:37 ` Qu Wenruo
2026-02-24  4:42   ` Qu Wenruo
2026-02-24 13:52     ` David Sterba
2026-02-24  6:36   ` Miquel Sabaté Solà
     [not found]   ` <699d4704.050a0220.1a6450.86d7SMTPIN_ADDED_BROKEN@mx.google.com>
2026-02-24  6:48     ` Qu Wenruo
2026-02-24  8:59       ` Miquel Sabaté Solà [this message]
2026-02-24 14:55   ` David Laight
2026-02-25 14:44     ` David Sterba
2026-02-25 17:11       ` David Laight
2026-02-24  6:32 ` Johannes Thumshirn
2026-02-24  6:46   ` Miquel Sabaté Solà
2026-02-24  6:54     ` Qu Wenruo
2026-02-24  9:04       ` Miquel Sabaté Solà

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87qzqa5xgn.fsf@ \
    --to=mssola@mssola.com \
    --cc=clm@fb.com \
    --cc=dsterba@suse.com \
    --cc=kees@kernel.org \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=naohiro.aota@wdc.com \
    --cc=quwenruo.btrfs@gmx.com \
    --cc=wqu@suse.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox