linux-kernel - Re: [PATCH] interconnect: Check for valid path in icc_set

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: <20191220190426.GE549437@yoga>
Date:   Fri, 20 Dec 2019 11:04:26 -0800
From:   Bjorn Andersson <bjorn.andersson@...aro.org>
To:     Georgi Djakov <georgi.djakov@...aro.org>
Cc:     linux-pm@...r.kernel.org, evgreen@...omium.org,
        daidavid1@...eaurora.org, okukatla@...eaurora.org,
        jcrouse@...eaurora.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] interconnect: Check for valid path in icc_set_bw()

On Fri 20 Dec 09:13 PST 2019, Georgi Djakov wrote:

> Use IS_ERR() to ensure that the path passed to icc_set_bw() is valid.
> 
> Signed-off-by: Georgi Djakov <georgi.djakov@...aro.org>
> ---
>  drivers/interconnect/core.c | 5 +++++
>  1 file changed, 5 insertions(+)
> 
> diff --git a/drivers/interconnect/core.c b/drivers/interconnect/core.c
> index 63c164264b73..14a6f7ade44a 100644
> --- a/drivers/interconnect/core.c
> +++ b/drivers/interconnect/core.c
> @@ -498,6 +498,11 @@ int icc_set_bw(struct icc_path *path, u32 avg_bw, u32 peak_bw)
>  	if (!path || !path->num_nodes)
>  		return 0;
>  
> +	if (IS_ERR(path)) {

This is a sign of a logical error, and the print is likely to be
ignored/lost in the noise. So I think the response should aid to help
the developer hitting this to resolve the issue.

So I think this is more visible and more useful as:

	if (WARN_ON(IS_ERR(path)))
		return -EINVAL;


PS. Doesn't path->num_nodes == 0 fall in this category as well? When
would you have a path object with no nodes passed to this function?

Regards,
Bjorn

> +		pr_err("%s: invalid path=%ld\n", __func__, PTR_ERR(path));
> +		return -EINVAL;
> +	}
> +
>  	mutex_lock(&icc_lock);
>  
>  	old_avg = path->reqs[0].avg_bw;