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]
Message-ID: <6d8ac0cf-b954-4c12-8b5b-e172c850e529@intel.com>
Date: Tue, 9 Jul 2024 10:02:05 +0200
From: Przemek Kitszel <przemyslaw.kitszel@...el.com>
To: Aleksandr Mishin <amishin@...rgos.ru>
CC: Jesse Brandeburg <jesse.brandeburg@...el.com>, Tony Nguyen
	<anthony.l.nguyen@...el.com>, "David S. Miller" <davem@...emloft.net>, "Eric
 Dumazet" <edumazet@...gle.com>, Jakub Kicinski <kuba@...nel.org>, Paolo Abeni
	<pabeni@...hat.com>, <intel-wired-lan@...ts.osuosl.org>,
	<netdev@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
	<lvc-project@...uxtesting.org>
Subject: Re: [PATCH net-next v3] ice: Adjust over allocation of memory in
 ice_sched_add_root_node() and ice_sched_add_node()

On 7/8/24 20:27, Aleksandr Mishin wrote:
> In ice_sched_add_root_node() and ice_sched_add_node() there are calls to
> devm_kcalloc() in order to allocate memory for array of pointers to
> 'ice_sched_node' structure. But incorrect types are used as sizeof()
> arguments in these calls (structures instead of pointers) which leads to
> over allocation of memory.
> 
> Adjust over allocation of memory by correcting types in devm_kcalloc()
> sizeof() arguments.
> 
> Found by Linux Verification Center (linuxtesting.org) with SVACE.
> 
> Suggested-by: Przemek Kitszel <przemyslaw.kitszel@...el.com>

Suggested-by tag is for when the Suggester is the raison d'etre of the
commit, not just improved it via review.
Here there is no need for this tag.

As an example:
We don't like the current state with the abuse of devm_ family of
allocations in the ice driver, those are used in places where plain
kzalloc() will fit better.

With the previous paragraph you could go to review our driver and swap
devm_kzalloc() for kzalloc() in some cases, "proving" that this is
a correct change; and finally add my Suggested-by. You could've done
the same without the tag if you will instead assume that this suggestion
was too obvious or too broad or too trivial, but it is just an example:)

> Signed-off-by: Aleksandr Mishin <amishin@...rgos.ru>

I like the new code, so:
Reviewed-by: Przemek Kitszel <przemyslaw.kitszel@...el.com>

> ---
> v3:
>    - Update comment and use the correct entities as suggested by Przemek
> v2: https://lore.kernel.org/all/20240706140518.9214-1-amishin@t-argos.ru/
>    - Update comment, remove 'Fixes' tag and change the tree from 'net' to
>      'net-next' as suggested by Simon
>      (https://lore.kernel.org/all/20240706095258.GB1481495@kernel.org/)
> v1: https://lore.kernel.org/all/20240705163620.12429-1-amishin@t-argos.ru/
> 
>   drivers/net/ethernet/intel/ice/ice_sched.c | 6 ++----
>   1 file changed, 2 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/net/ethernet/intel/ice/ice_sched.c b/drivers/net/ethernet/intel/ice/ice_sched.c
> index ecf8f5d60292..6ca13c5dcb14 100644
> --- a/drivers/net/ethernet/intel/ice/ice_sched.c
> +++ b/drivers/net/ethernet/intel/ice/ice_sched.c
> @@ -28,9 +28,8 @@ ice_sched_add_root_node(struct ice_port_info *pi,
>   	if (!root)
>   		return -ENOMEM;
>   
> -	/* coverity[suspicious_sizeof] */
>   	root->children = devm_kcalloc(ice_hw_to_dev(hw), hw->max_children[0],
> -				      sizeof(*root), GFP_KERNEL);
> +				      sizeof(*root->children), GFP_KERNEL);
>   	if (!root->children) {
>   		devm_kfree(ice_hw_to_dev(hw), root);
>   		return -ENOMEM;
> @@ -186,10 +185,9 @@ ice_sched_add_node(struct ice_port_info *pi, u8 layer,
>   	if (!node)
>   		return -ENOMEM;
>   	if (hw->max_children[layer]) {
> -		/* coverity[suspicious_sizeof] */
>   		node->children = devm_kcalloc(ice_hw_to_dev(hw),
>   					      hw->max_children[layer],
> -					      sizeof(*node), GFP_KERNEL);
> +					      sizeof(*node->children), GFP_KERNEL);
>   		if (!node->children) {
>   			devm_kfree(ice_hw_to_dev(hw), node);
>   			return -ENOMEM;


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ