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] [day] [month] [year] [list]
Message-ID: <16df2125-fdf3-49e9-9924-425a8d7e1377@kernel.dk>
Date: Thu, 14 Mar 2024 13:29:09 -0600
From: Jens Axboe <axboe@...nel.dk>
To: Xin Wang <yw987194828@...il.com>
Cc: asml.silence@...il.com, io-uring@...r.kernel.org,
 linux-kernel@...r.kernel.org, Xin Wang <yw987194828@....com>
Subject: Re: [PATCH] io_uring: extract the function that checks the legitimacy
 of sq/cq entries

On 3/12/24 1:44 PM, Xin Wang wrote:
> In the io_uring_create function, the sq_entries and cq_entries passed
> in by the user are examined. The checking logic is the same for both, so
> the common code can be extracted for reuse.

Looks fine to me, though not sure how helpful it really is, it's not
like it's a lot of code and it's easy enough to read as it is. However,
a few minor comments:

>  					 O_RDWR | O_CLOEXEC, NULL);
>  }
>  
> +static bool io_validate_entries(unsigned int *entries, unsigned int max_entries, __u32 flags)

Line too long, please break list other functions. Also needs a better
name, probably io_validate_ring_entries() would be better.

> +{
> +	if (!(*entries))
> +		return false;
> +	if (*entries > max_entries) {
> +		if (!(flags & IORING_SETUP_CLAMP))
> +			return false;
> +		*entries = max_entries;
> +	}
> +	return true;
> +}

And I don't know why you use parens for the first *entries check, but
then not for the next? Should be consistent, at least.

> @@ -3854,13 +3861,8 @@ static __cold int io_uring_create(unsigned entries, struct io_uring_params *p,
>  		 * to a power-of-two, if it isn't already. We do NOT impose
>  		 * any cq vs sq ring sizing.
>  		 */
> -		if (!p->cq_entries)
> +		if (!io_validate_entries(&(p->cq_entries), IORING_MAX_CQ_ENTRIES, p->flags))

Again not sure what these parens are doing here?

-- 
Jens Axboe


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ