| 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
| ||
|
Date: Tue, 15 Aug 2023 19:56:16 -0400
From: Waiman Long <longman@...hat.com>
To: kernel test robot <lkp@...el.com>, Tejun Heo <tj@...nel.org>,
Zefan Li <lizefan.x@...edance.com>,
Johannes Weiner <hannes@...xchg.org>,
Christian Brauner <brauner@...nel.org>,
Jonathan Corbet <corbet@....net>,
Shuah Khan <skhan@...uxfoundation.org>
Cc: oe-kbuild-all@...ts.linux.dev, cgroups@...r.kernel.org,
linux-kernel@...r.kernel.org, Juri Lelli <juri.lelli@...hat.com>,
Dietmar Eggemann <dietmar.eggemann@....com>,
Michal Koutný <mkoutny@...e.com>,
Giuseppe Scrivano <gscrivan@...hat.com>
Subject: Re: [PATCH-cgroup v6 1/6] cgroup/cpuset: Add
cpuset.cpus.exclusive.effective for v2
On 8/15/23 19:18, kernel test robot wrote:
> Hi Waiman,
>
> kernel test robot noticed the following build warnings:
>
> [auto build test WARNING on next-20230815]
> [cannot apply to tj-cgroup/for-next v6.5-rc6 v6.5-rc5 v6.5-rc4 linus/master v6.5-rc6]
> [If your patch is applied to the wrong git tree, kindly drop us a note.
> And when submitting patch, we suggest to use '--base' as documented in
> https://git-scm.com/docs/git-format-patch#_base_tree_information]
>
> url: https://github.com/intel-lab-lkp/linux/commits/Waiman-Long/cgroup-cpuset-Add-cpuset-cpus-exclusive-effective-for-v2/20230815-233522
> base: next-20230815
> patch link: https://lore.kernel.org/r/20230815153027.633355-2-longman%40redhat.com
> patch subject: [PATCH-cgroup v6 1/6] cgroup/cpuset: Add cpuset.cpus.exclusive.effective for v2
> config: parisc64-defconfig (https://download.01.org/0day-ci/archive/20230816/202308160752.IWyhJoln-lkp@intel.com/config)
> compiler: hppa-linux-gcc (GCC) 12.3.0
> reproduce: (https://download.01.org/0day-ci/archive/20230816/202308160752.IWyhJoln-lkp@intel.com/reproduce)
>
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@...el.com>
> | Closes: https://lore.kernel.org/oe-kbuild-all/202308160752.IWyhJoln-lkp@intel.com/
>
> All warnings (new ones prefixed by >>):
>
> kernel/cgroup/cpuset.c: In function 'update_prstate':
>>> kernel/cgroup/cpuset.c:2468:30: warning: the comparison will always evaluate as 'true' for the address of 'effective_xcpus' will never be NULL [-Waddress]
> 2468 | if ((new_prs > 0) && !cs->effective_xcpus) {
Thanks for catching that. It is a careless mistake. Will send out a fix
version.
Cheers,
Longman
> | ^
> kernel/cgroup/cpuset.c:135:23: note: 'effective_xcpus' declared here
> 135 | cpumask_var_t effective_xcpus;
> | ^~~~~~~~~~~~~~~
>
>
> vim +2468 kernel/cgroup/cpuset.c
>
> 2435
> 2436 /**
> 2437 * update_prstate - update partition_root_state
> 2438 * @cs: the cpuset to update
> 2439 * @new_prs: new partition root state
> 2440 * Return: 0 if successful, != 0 if error
> 2441 *
> 2442 * Call with cpuset_mutex held.
> 2443 */
> 2444 static int update_prstate(struct cpuset *cs, int new_prs)
> 2445 {
> 2446 int err = PERR_NONE, old_prs = cs->partition_root_state;
> 2447 struct tmpmasks tmpmask;
> 2448
> 2449 if (old_prs == new_prs)
> 2450 return 0;
> 2451
> 2452 /*
> 2453 * For a previously invalid partition root, leave it at being
> 2454 * invalid if new_prs is not "member".
> 2455 */
> 2456 if (new_prs && is_prs_invalid(old_prs)) {
> 2457 cs->partition_root_state = -new_prs;
> 2458 return 0;
> 2459 }
> 2460
> 2461 if (alloc_cpumasks(NULL, &tmpmask))
> 2462 return -ENOMEM;
> 2463
> 2464 /*
> 2465 * Setup effective_xcpus if not set yet, it will be cleared later
> 2466 * if partition becomes invalid.
> 2467 */
>> 2468 if ((new_prs > 0) && !cs->effective_xcpus) {
> 2469 struct cpuset *parent = parent_cs(cs);
> 2470
> 2471 spin_lock_irq(&callback_lock);
> 2472 cpumask_and(cs->effective_xcpus,
> 2473 cs->cpus_allowed, parent->effective_xcpus);
> 2474 spin_unlock_irq(&callback_lock);
> 2475 }
> 2476
> 2477 err = update_partition_exclusive(cs, new_prs);
> 2478 if (err)
> 2479 goto out;
> 2480
> 2481 if (!old_prs) {
> 2482 /*
> 2483 * cpus_allowed cannot be empty.
> 2484 */
> 2485 if (cpumask_empty(cs->cpus_allowed)) {
> 2486 err = PERR_CPUSEMPTY;
> 2487 goto out;
> 2488 }
> 2489
> 2490 err = update_parent_effective_cpumask(cs, partcmd_enable,
> 2491 NULL, &tmpmask);
> 2492 } else if (old_prs && new_prs) {
> 2493 /*
> 2494 * A change in load balance state only, no change in cpumasks.
> 2495 */
> 2496 ;
> 2497 } else {
> 2498 /*
> 2499 * Switching back to member is always allowed even if it
> 2500 * disables child partitions.
> 2501 */
> 2502 update_parent_effective_cpumask(cs, partcmd_disable, NULL,
> 2503 &tmpmask);
> 2504
> 2505 /*
> 2506 * Invalidation of child partitions will be done in
> 2507 * update_cpumasks_hier().
> 2508 */
> 2509 }
> 2510 out:
> 2511 /*
> 2512 * Make partition invalid & disable CS_CPU_EXCLUSIVE if an error
> 2513 * happens.
> 2514 */
> 2515 if (err) {
> 2516 new_prs = -new_prs;
> 2517 update_partition_exclusive(cs, new_prs);
> 2518 }
> 2519
> 2520 spin_lock_irq(&callback_lock);
> 2521 cs->partition_root_state = new_prs;
> 2522 WRITE_ONCE(cs->prs_err, err);
> 2523 if (!is_partition_valid(cs))
> 2524 cpumask_clear(cs->effective_xcpus);
> 2525 spin_unlock_irq(&callback_lock);
> 2526
> 2527 /* Force update if switching back to member */
> 2528 update_cpumasks_hier(cs, &tmpmask, !new_prs ? HIER_CHECKALL : 0);
> 2529
> 2530 /* Update sched domains and load balance flag */
> 2531 update_partition_sd_lb(cs, old_prs);
> 2532
> 2533 notify_partition_change(cs, old_prs);
> 2534 free_cpumasks(NULL, &tmpmask);
> 2535 return 0;
> 2536 }
> 2537
>
Powered by blists - more mailing lists