linux-kernel - Re: [PATCH next] cgroup/cpuset: Cleanup signedness issue in cpu_exclusive

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: <aceadbf1-0345-2b7d-4468-c7c42ac15e03@oracle.com>
Date:   Wed, 27 Sep 2023 15:07:02 +0530
From:   Kamalesh Babulal <kamalesh.babulal@...cle.com>
To:     Harshit Mogalapalli <harshit.m.mogalapalli@...cle.com>,
        Waiman Long <longman@...hat.com>,
        Zefan Li <lizefan.x@...edance.com>, Tejun Heo <tj@...nel.org>,
        Johannes Weiner <hannes@...xchg.org>, cgroups@...r.kernel.org,
        linux-kernel@...r.kernel.org
Cc:     dan.carpenter@...aro.org, kernel-janitors@...r.kernel.org,
        error27@...il.com, kernel test robot <lkp@...el.com>
Subject: Re: [PATCH next] cgroup/cpuset: Cleanup signedness issue in
 cpu_exclusive_check()


On 9/27/23 12:28, Harshit Mogalapalli wrote:
> Smatch complains about returning negative error codes from a type
> bool function.
> 
> kernel/cgroup/cpuset.c:705 cpu_exclusive_check() warn:
> 	signedness bug returning '(-22)'
> 
> The code works correctly, but it is confusing.  The current behavior is
> that cpu_exclusive_check() returns true if it's *NOT* exclusive.  Rename
> it to cpusets_are_exclusive() and reverse the returns so it returns true
> if it is exclusive and false if it's not.  Update both callers as well.
> 
> Reported-by: kernel test robot <lkp@...el.com>
> Reported-by: Dan Carpenter <error27@...il.com>
> Closes: https://lore.kernel.org/r/202309201706.2LhKdM6o-lkp@intel.com/
> Signed-off-by: Harshit Mogalapalli <harshit.m.mogalapalli@...cle.com>

The patch looks good to me, returning true on exclusive cpusets is more
intuitive. Renaming cpu_exclusive_check() to is_cpuset_exclusive() is
one other option, though, there is is_cpu_exclusive() function, which
sounds similar, and tests for the cpu exclusive bit in a given cpuset's
flag.

I don't have a strong opinion between the original function name and
the proposed rename.

Reviewed-by: Kamalesh Babulal <kamalesh.babulal@...cle.com>

> ---
> This is based on sattic analysis, only compile tested
> ---
>  kernel/cgroup/cpuset.c | 14 +++++++-------
>  1 file changed, 7 insertions(+), 7 deletions(-)
> 
> diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
> index 15f399153a2e..afefddd33c3e 100644
> --- a/kernel/cgroup/cpuset.c
> +++ b/kernel/cgroup/cpuset.c
> @@ -719,18 +719,18 @@ static inline struct cpumask *fetch_xcpus(struct cpuset *cs)
>  }
>  
>  /*
> - * cpu_exclusive_check() - check if two cpusets are exclusive
> + * cpusets_are_exclusive() - check if two cpusets are exclusive
>   *
> - * Return 0 if exclusive, -EINVAL if not
> + * Return true if exclusive, false if not
>   */
> -static inline bool cpu_exclusive_check(struct cpuset *cs1, struct cpuset *cs2)
> +static inline bool cpusets_are_exclusive(struct cpuset *cs1, struct cpuset *cs2)
>  {
>  	struct cpumask *xcpus1 = fetch_xcpus(cs1);
>  	struct cpumask *xcpus2 = fetch_xcpus(cs2);
>  
>  	if (cpumask_intersects(xcpus1, xcpus2))
> -		return -EINVAL;
> -	return 0;
> +		return false;
> +	return true;
>  }
>  
>  /*
> @@ -833,7 +833,7 @@ static int validate_change(struct cpuset *cur, struct cpuset *trial)
>  	cpuset_for_each_child(c, css, par) {
>  		if ((is_cpu_exclusive(trial) || is_cpu_exclusive(c)) &&
>  		    c != cur) {
> -			if (cpu_exclusive_check(trial, c))
> +			if (!cpusets_are_exclusive(trial, c))
>  				goto out;
>  		}
>  		if ((is_mem_exclusive(trial) || is_mem_exclusive(c)) &&
> @@ -1864,7 +1864,7 @@ static int update_parent_effective_cpumask(struct cpuset *cs, int cmd,
>  			cpuset_for_each_child(child, css, parent) {
>  				if (child == cs)
>  					continue;
> -				if (cpu_exclusive_check(cs, child)) {
> +				if (!cpusets_are_exclusive(cs, child)) {
>  					exclusive = false;
>  					break;
>  				}

-- 
Thanks,
Kamalesh