linux-kernel - Re: [PATCH v2 04/16] fs: Increase fmode

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 13 Dec 2023 12:20:26 +0100
From:   Jan Kara <jack@...e.cz>
To:     John Garry <john.g.garry@...cle.com>
Cc:     axboe@...nel.dk, kbusch@...nel.org, hch@....de, sagi@...mberg.me,
        jejb@...ux.ibm.com, martin.petersen@...cle.com, djwong@...nel.org,
        viro@...iv.linux.org.uk, brauner@...nel.org, dchinner@...hat.com,
        jack@...e.cz, linux-block@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-nvme@...ts.infradead.org,
        linux-xfs@...r.kernel.org, linux-fsdevel@...r.kernel.org,
        tytso@....edu, jbongio@...gle.com, linux-scsi@...r.kernel.org,
        ming.lei@...hat.com, jaswin@...ux.ibm.com, bvanassche@....org
Subject: Re: [PATCH v2 04/16] fs: Increase fmode_t size

On Tue 12-12-23 11:08:32, John Garry wrote:
> Currently all bits are being used in fmode_t.
> 
> To allow for further expansion, increase from unsigned int to unsigned
> long.
> 
> Since the dma-buf driver prints the file->f_mode member, change the print
> as necessary to deal with the larger size.
> 
> Signed-off-by: John Garry <john.g.garry@...cle.com>

Uh, Al has more experience with fmode_t changes so I'd defer final decision
to him but to me this seems dangerous. Firstly, this breaks packing of
struct file on 64-bit architectures and struct file is highly optimized for
cache efficiency (see the comment before the struct definition). Secondly
this will probably generate warnings on 32-bit architectures as there
sizeof(unsigned long) == sizeof(unsigned int) and so your new flags won't
fit anyway?

								Honza

> ---
>  drivers/dma-buf/dma-buf.c | 2 +-
>  include/linux/types.h     | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c
> index 21916bba77d5..a5227ae3d637 100644
> --- a/drivers/dma-buf/dma-buf.c
> +++ b/drivers/dma-buf/dma-buf.c
> @@ -1628,7 +1628,7 @@ static int dma_buf_debug_show(struct seq_file *s, void *unused)
>  
>  
>  		spin_lock(&buf_obj->name_lock);
> -		seq_printf(s, "%08zu\t%08x\t%08x\t%08ld\t%s\t%08lu\t%s\n",
> +		seq_printf(s, "%08zu\t%08x\t%08lx\t%08ld\t%s\t%08lu\t%s\n",
>  				buf_obj->size,
>  				buf_obj->file->f_flags, buf_obj->file->f_mode,
>  				file_count(buf_obj->file),
> diff --git a/include/linux/types.h b/include/linux/types.h
> index 253168bb3fe1..49c754fde1d6 100644
> --- a/include/linux/types.h
> +++ b/include/linux/types.h
> @@ -153,7 +153,7 @@ typedef u32 dma_addr_t;
>  
>  typedef unsigned int __bitwise gfp_t;
>  typedef unsigned int __bitwise slab_flags_t;
> -typedef unsigned int __bitwise fmode_t;
> +typedef unsigned long __bitwise fmode_t;
>  
>  #ifdef CONFIG_PHYS_ADDR_T_64BIT
>  typedef u64 phys_addr_t;
> -- 
> 2.35.3
> 
-- 
Jan Kara <jack@...e.com>
SUSE Labs, CR