Re: [PATCH v5 1/6] cgroup/cpuset: Properly transition to invalid partition

From: kernel test robot
Date: Sat Aug 14 2021 - 16:22:26 EST

Next message:
Previous message:
In reply to:
Next in thread: Messages sorted by:
Hi Waiman,

I love your patch! Perhaps something to
[auto build test WARNING on cgroup/for-next]
[also build test WARNING on next-20210813]
[cannot apply to kselftest/next v5.14-rc5]
[If your patch is applied to the wrong git And when submitting patch, we suggest to .config.gzDescription: application/gzip kernel test robot: "arch/arm64/kvm/va_layout.c:292:6: warning: no previous prototype for function 'kvm_compute_final_ctr_el0'" Christophe JAILLET: "Re: [PATCH] checkpatch: prefer = {} initializations to = {0}" Waiman Long: "[PATCH v5 1/6] cgroup/cpuset: Properly transition to invalid partition" href="07320.html"> Waiman Long: "Re: [PATCH v5 1/6] cgroup/cpuset: Properly transition to invalid partition" href="date.html#07316">[ date ] [ thread ] [ subject ] [ author ] improve:
tree, kindly drop us a note.
use '--base' as documented in
/git-format-patch">https://git-scm.com/docs/git-format-patch]
om/0day-ci/linux/commits/Waiman-Long/cgroup-cpuset-Add-new-cpuset-partition-type-empty-effecitve-cpus/20210815-014333">https://github.com/0day-ci/linux/commits/Waiman-Long/cgroup-cpuset-Add-new-cpuset-partition-type-empty-effecitve-cpus/20210815-014333
el.org/pub/scm/linux/kernel/git/tj/cgroup.git">https://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup.git for-next
.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross">https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
.com/0day-ci/linux/commit/56ec7dd271c77e3cc92f0df6fd766004a7a0aa88">https://github.com/0day-ci/linux/commit/56ec7dd271c77e3cc92f0df6fd766004a7a0aa88
rel="nofollow" href="https://github.com/0day-ci/linux">https://github.com/0day-ci/linux
Waiman-Long/cgroup-cpuset-Add-new-cpuset-partition-type-empty-effecitve-cpus/20210815-014333
6004a7a0aa88
build tree
COMPILER=gcc-11.2.0 make.cross ARCH=ia64
tag as appropriate
gt;
state':
warning: the frame size of 3072 bytes is larger than 2048 bytes [-Wframe-larger-than=]
Torvalds 2005-04-16 1966
2018-11-08 1967 /*
2018-11-08 1968 * update_prstate - update partititon_root_state
2018-11-08 1969 * cs: the cpuset to update
2021-07-20 1970 * new_prs: new partition root state
2018-11-08 1971 *
2018-11-08 1972 * Call with cpuset_mutex held.
2018-11-08 1973 */
2021-07-20 1974 static int update_prstate(struct cpuset *cs, int new_prs)
2018-11-08 1975 {
2021-07-20 1976 int err, old_prs = cs->partition_root_state;
2018-11-08 1977 struct cpuset *parent = parent_cs(cs);
2021-07-20 1978 struct tmpmasks tmpmask;
2018-11-08 1979
2021-07-20 1980 if (old_prs == new_prs)
2018-11-08 1981 return 0;
2018-11-08 1982
2018-11-08 1983 /*
2018-11-08 1984 * Cannot force a partial or invalid partition root to a full
2018-11-08 1985 * partition root.
2018-11-08 1986 */
2021-07-20 1987 if (new_prs && (old_prs == PRS_ERROR))
2018-11-08 1988 return -EINVAL;
2018-11-08 1989
2021-07-20 1990 if (alloc_cpumasks(NULL, &tmpmask))
2018-11-08 1991 return -ENOMEM;
2018-11-08 1992
2018-11-08 1993 err = -EINVAL;
2021-07-20 1994 if (!old_prs) {
2018-11-08 1995 /*
2018-11-08 1996 * Turning on partition root requires setting the
2018-11-08 1997 * CS_CPU_EXCLUSIVE bit implicitly as well and cpus_allowed
2018-11-08 1998 * cannot be NULL.
2018-11-08 1999 */
2018-11-08 2000 if (cpumask_empty(cs->cpus_allowed))
2018-11-08 2001 goto out;
2018-11-08 2002
2018-11-08 2003 err = update_flag(CS_CPU_EXCLUSIVE, cs, 1);
2018-11-08 2004 if (err)
2018-11-08 2005 goto out;
2018-11-08 2006
2018-11-08 2007 err = update_parent_subparts_cpumask(cs, partcmd_enable,
2021-07-20 2008 NULL, &tmpmask);
2018-11-08 2009 if (err) {
2018-11-08 2010 update_flag(CS_CPU_EXCLUSIVE, cs, 0);
2018-11-08 2011 goto out;
2018-11-08 2012 }
2018-11-08 2013 } else {
2018-11-08 2014 /*
2021-08-14 2015 * Switch back to member is always allowed even if it
2021-08-14 2016 * causes child partitions to become invalid.
2018-11-08 2017 */
2018-11-08 2018 err = 0;
2021-08-14 2019 update_parent_subparts_cpumask(cs, partcmd_disable, NULL,
2021-08-14 2020 &tmpmask);
2021-08-14 2021 /*
2021-08-14 2022 * If there are child partitions, we have to make them invalid.
2021-08-14 2023 */
2021-08-14 2024 if (unlikely(cs->nr_subparts_cpus)) {
2021-08-14 2025 struct tmpmasks tmp;
2018-11-08 2026
2021-08-14 2027 spin_lock_irq(&callback_lock);
2021-08-14 2028 cs->nr_subparts_cpus = 0;
2021-08-14 2029 cpumask_clear(cs->subparts_cpus);
2021-08-14 2030 compute_effective_cpumask(cs->effective_cpus, cs, parent);
2021-08-14 2031 spin_unlock_irq(&callback_lock);
2021-08-14 2032
2021-08-14 2033 /*
2021-08-14 2034 * If alloc_cpumasks() fails, we are running out
2021-08-14 2035 * of memory and there isn't much we can do.
2021-08-14 2036 */
2021-08-14 2037 if (!alloc_cpumasks(NULL, &tmp)) {
2021-08-14 2038 update_cpumasks_hier(cs, &tmp);
2021-08-14 2039 free_cpumasks(NULL, &tmp);
2021-08-14 2040 }
2021-08-14 2041 }
2018-11-08 2042
2018-11-08 2043 /* Turning off CS_CPU_EXCLUSIVE will not return error */
2018-11-08 2044 update_flag(CS_CPU_EXCLUSIVE, cs, 0);
2018-11-08 2045 }
2018-11-08 2046
2018-11-08 2047 /*
2018-11-08 2048 * Update cpumask of parent's tasks except when it is the top
2018-11-08 2049 * cpuset as some system daemons cannot be mapped to other CPUs.
2018-11-08 2050 */
2018-11-08 2051 if (parent != &top_cpuset)
2018-11-08 2052 update_tasks_cpumask(parent);
2018-11-08 2053
2018-11-08 2054 if (parent->child_ecpus_count)
2021-07-20 2055 update_sibling_cpumasks(parent, cs, &tmpmask);
2018-11-08 2056
2018-11-08 2057 rebuild_sched_domains_locked();
2018-11-08 2058 out:
2021-07-20 2059 if (!err) {
2021-07-20 2060 spin_lock_irq(&callback_lock);
2021-07-20 2061 cs->partition_root_state = new_prs;
2021-07-20 2062 spin_unlock_irq(&callback_lock);
2021-08-10 2063 notify_partition_change(cs, old_prs, new_prs);
2021-07-20 2064 }
2021-07-20 2065
2021-07-20 2066 free_cpumasks(NULL, &tmpmask);
2009-01-07 2067 return err;
Torvalds 2005-04-16 @2068 }
Torvalds 2005-04-16 2069
r> erkitty/list/kbuild-all@xxxxxxxxxxxx">https://lists.01.org/hyperkitty/list/kbuild-all@xxxxxxxxxxxx
a>

Next message: kernel test robot: "arch/arm64/kvm/va_layout.c:292:6: warning: no previous prototype for function 'kvm_compute_final_ctr_el0'"
Previous message: Christophe JAILLET: "Re: [PATCH] checkpatch: prefer = {} initializations to = {0}"
In reply to: Waiman Long: "[PATCH v5 1/6] cgroup/cpuset: Properly transition to invalid partition"
Next in thread: Waiman Long: "Re: [PATCH v5 1/6] cgroup/cpuset: Properly transition to invalid partition"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]