Re: [PATCH] scsi: Use kmalloc_array to prevent overflow of dynamic size calculation

[James Bottomley <James.Bottomley@HansenPartnership.com>]

On Wed, 2025-10-01 at 17:09 +0530, Bhanu Seshu Kumar Valluri wrote:
> Use kmalloc_array to avoid potential overflow during dynamic size
> calculation inside kmalloc.

This description isn't correct.

> 
> Signed-off-by: Bhanu Seshu Kumar Valluri <bhanuseshukumar@gmail.com>
> ---
>  drivers/scsi/smartpqi/smartpqi_init.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/scsi/smartpqi/smartpqi_init.c
> b/drivers/scsi/smartpqi/smartpqi_init.c
> index 125944941601..7ff39f1faf38 100644
> --- a/drivers/scsi/smartpqi/smartpqi_init.c
> +++ b/drivers/scsi/smartpqi/smartpqi_init.c
> @@ -8937,7 +8937,7 @@ static int pqi_host_alloc_mem(struct
> pqi_ctrl_info *ctrl_info,
>  	if (sg_count == 0 || sg_count > PQI_HOST_MAX_SG_DESCRIPTORS)
>  		goto out;

Given this check

>  
> -	host_memory_descriptor->host_chunk_virt_address =
> kmalloc(sg_count * sizeof(void *), GFP_KERNEL);

How is it possible that this allocation could ever overflow?

If you want to change the description to say using kmalloc_array is
better practice or something (and the maintainer concurs) that's fine,
but we can't have a false justification in the kernel git log.

Regards,

James