linux-kernel - Re: [PATCH v2] cpupower:Fix resource leaks in sysfs_get

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:   Fri, 14 Apr 2023 10:41:35 -0600
From:   Shuah Khan <skhan@...uxfoundation.org>
To:     Hao Zeng <zenghao@...inos.cn>
Cc:     trenn@...e.com, shuah@...nel.org, linux-pm@...r.kernel.org,
        linux-kernel@...r.kernel.org,
        Shuah Khan <skhan@...uxfoundation.org>
Subject: Re: [PATCH v2] cpupower:Fix resource leaks in sysfs_get_enabled()

On 4/13/23 21:08, Hao Zeng wrote:
> When the read return value is equal to 1, a file handle leak will occur
> 

Would like a bit more information on how the error paths are
redone, since memory leak isn't the only one that is fixed.

> Signed-off-by: Hao Zeng <zenghao@...inos.cn>
> Suggested-by: Shuah Khan <skhan@...uxfoundation.org>
> ---
>   tools/power/cpupower/lib/powercap.c | 19 ++++++++++---------
>   1 file changed, 10 insertions(+), 9 deletions(-)
> 
> diff --git a/tools/power/cpupower/lib/powercap.c b/tools/power/cpupower/lib/powercap.c
> index 0ce29ee4c2e4..02ec5b0bff6b 100644
> --- a/tools/power/cpupower/lib/powercap.c
> +++ b/tools/power/cpupower/lib/powercap.c
> @@ -40,7 +40,7 @@ static int sysfs_get_enabled(char *path, int *mode)
>   {
>   	int fd;
>   	char yes_no;
> -
> +	int ret = 0;
>   	*mode = 0;
>   
>   	fd = open(path, O_RDONLY);
> @@ -48,17 +48,18 @@ static int sysfs_get_enabled(char *path, int *mode)
>   		return -1;
>   
>   	if (read(fd, &yes_no, 1) != 1) {
> -		close(fd);
> -		return -1;
> +		ret = -1;
> +		goto err;
>   	}
>   
> -	if (yes_no == '1') {
> -		*mode = 1;
> -		return 0;
> -	} else if (yes_no == '0') {
> -		return 0;
> +	if (yes_no != '1' || yes_no != '0') {
> +		ret = -1;
> +		goto err;
>   	}
> -	return -1;
> +	*mode = yes_no - '0';

I am not seeing much value in changing the above paths.
Leave them unchanged.

> +err:
> +	close(fd);
> +	return ret;
>   }
>   
>   int powercap_get_enabled(int *mode)

thanks,
-- Shuah