* [PATCH 0/4] Use __counted_by for ancestor arrays @ 2025-12-17 16:27 Michal Koutný 2025-12-17 16:27 ` [PATCH 2/4] cgroup: Introduce cgroup_level() helper Michal Koutný 0 siblings, 1 reply; 5+ messages in thread From: Michal Koutný @ 2025-12-17 16:27 UTC (permalink / raw) To: linux-block, bpf, linux-trace-kernel, netfilter-devel, netdev, coreteam, linux-hardening, linux-kernel, cgroups Cc: Michal Koutný, Hao Luo, Mathieu Desnoyers, Alexei Starovoitov, Phil Sutter, Yonghong Song, Jens Axboe, Jozsef Kadlecsik, Steven Rostedt, Martin KaFai Lau, KP Singh, Eric Dumazet, Florian Westphal, Jiri Olsa, Stanislav Fomichev, Song Liu, David S. Miller, Simon Horman, John Fastabend, Johannes Weiner, Daniel Borkmann, Andrii Nakryiko, Tejun Heo, Josef Bacik, Paolo Abeni, Gustavo A. R. Silva, Pablo Neira Ayuso, Eduard Zingerman, Yu Kuai, Masami Hiramatsu, Kees Cook, Jakub Kicinski The trick with utilizing space in cgroup_root for cgroup::ancetors flex array was an obstacle for kernel reworks for -Wflex-array-member-not-at-end. The first patch fixes that, then I wanted to utilize __counted_by for this flex array which required some more rework how cgroup level is evaluated. Similar flex array is also in struct ioc_gq where it was tempting to simply use __counted_by(level), however, this would be off-by-one as it has semantics like cgroup's level (0 == root). Proper adjustment for __counted_by() would either need similar level/ancestor helpers or abstracted macros for ancestors up/down iterations. I only made a simple comment fixup since I'm not sure about benefit of __counted_by for structs that aren't sized based on direct user input. Michal Koutný (4): cgroup: Eliminate cgrp_ancestor_storage in cgroup_root cgroup: Introduce cgroup_level() helper cgroup: Use __counted_by for cgroup::ancestors blk-iocost: Correct comment ioc_gq::level block/bfq-iosched.c | 2 +- block/blk-iocost.c | 6 ++--- include/linux/cgroup-defs.h | 43 +++++++++++++++++++---------------- include/linux/cgroup.h | 18 ++++++++++++--- include/trace/events/cgroup.h | 8 +++---- kernel/bpf/helpers.c | 2 +- kernel/cgroup/cgroup.c | 9 ++++---- net/netfilter/nft_socket.c | 2 +- 8 files changed, 53 insertions(+), 37 deletions(-) base-commit: 8f0b4cce4481fb22653697cced8d0d04027cb1e8 -- 2.52.0 ^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH 2/4] cgroup: Introduce cgroup_level() helper 2025-12-17 16:27 [PATCH 0/4] Use __counted_by for ancestor arrays Michal Koutný @ 2025-12-17 16:27 ` Michal Koutný 2025-12-17 16:46 ` bot+bpf-ci ` (2 more replies) 0 siblings, 3 replies; 5+ messages in thread From: Michal Koutný @ 2025-12-17 16:27 UTC (permalink / raw) To: linux-block, linux-kernel, cgroups, linux-trace-kernel, bpf, netfilter-devel, coreteam, netdev Cc: Michal Koutný, Yu Kuai, Jens Axboe, Tejun Heo, Josef Bacik, Johannes Weiner, Steven Rostedt, Masami Hiramatsu, Mathieu Desnoyers, Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, Martin KaFai Lau, Eduard Zingerman, Song Liu, Yonghong Song, John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa, Pablo Neira Ayuso, Jozsef Kadlecsik, Florian Westphal, Phil Sutter, David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni, Simon Horman This is a no functional change to hide physical storage of cgroup's level and it allows subesequent conversion of the storage. Signed-off-by: Michal Koutný <mkoutny@suse.com> --- block/bfq-iosched.c | 2 +- block/blk-iocost.c | 4 ++-- include/linux/cgroup.h | 18 +++++++++++++++--- include/trace/events/cgroup.h | 8 ++++---- kernel/bpf/helpers.c | 2 +- kernel/cgroup/cgroup.c | 4 ++-- net/netfilter/nft_socket.c | 2 +- 7 files changed, 26 insertions(+), 14 deletions(-) diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c index 4a8d3d96bfe49..f293bab068274 100644 --- a/block/bfq-iosched.c +++ b/block/bfq-iosched.c @@ -601,7 +601,7 @@ static bool bfqq_request_over_limit(struct bfq_data *bfqd, goto out; /* +1 for bfqq entity, root cgroup not included */ - depth = bfqg_to_blkg(bfqq_group(bfqq))->blkcg->css.cgroup->level + 1; + depth = cgroup_level(bfqg_to_blkg(bfqq_group(bfqq))->blkcg->css.cgroup) + 1; if (depth > alloc_depth) { spin_unlock_irq(&bfqd->lock); if (entities != inline_entities) diff --git a/block/blk-iocost.c b/block/blk-iocost.c index a0416927d33dc..b4eebe61dca7f 100644 --- a/block/blk-iocost.c +++ b/block/blk-iocost.c @@ -2962,7 +2962,7 @@ static void ioc_cpd_free(struct blkcg_policy_data *cpd) static struct blkg_policy_data *ioc_pd_alloc(struct gendisk *disk, struct blkcg *blkcg, gfp_t gfp) { - int levels = blkcg->css.cgroup->level + 1; + int levels = cgroup_level(blkcg->css.cgroup) + 1; struct ioc_gq *iocg; iocg = kzalloc_node(struct_size(iocg, ancestors, levels), gfp, @@ -3003,7 +3003,7 @@ static void ioc_pd_init(struct blkg_policy_data *pd) init_waitqueue_head(&iocg->waitq); hrtimer_setup(&iocg->waitq_timer, iocg_waitq_timer_fn, CLOCK_MONOTONIC, HRTIMER_MODE_ABS); - iocg->level = blkg->blkcg->css.cgroup->level; + iocg->level = cgroup_level(blkg->blkcg->css.cgroup) for (tblkg = blkg; tblkg; tblkg = tblkg->parent) { struct ioc_gq *tiocg = blkg_to_iocg(tblkg); diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index bc892e3b37eea..0290878ebad26 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -525,6 +525,18 @@ static inline struct cgroup *cgroup_parent(struct cgroup *cgrp) return NULL; } +/** + * cgroup_level - cgroup depth + * @cgrp: cgroup + * + * The depth this cgroup is at. The root is at depth zero and each step down + * the hierarchy increments the level. + */ +static inline int cgroup_level(struct cgroup *cgrp) +{ + return cgrp->level; +} + /** * cgroup_is_descendant - test ancestry * @cgrp: the cgroup to be tested @@ -537,9 +549,9 @@ static inline struct cgroup *cgroup_parent(struct cgroup *cgrp) static inline bool cgroup_is_descendant(struct cgroup *cgrp, struct cgroup *ancestor) { - if (cgrp->root != ancestor->root || cgrp->level < ancestor->level) + if (cgrp->root != ancestor->root || cgroup_level(cgrp) < cgroup_level(ancestor)) return false; - return cgrp->ancestors[ancestor->level] == ancestor; + return cgrp->ancestors[cgroup_level(ancestor)] == ancestor; } /** @@ -556,7 +568,7 @@ static inline bool cgroup_is_descendant(struct cgroup *cgrp, static inline struct cgroup *cgroup_ancestor(struct cgroup *cgrp, int ancestor_level) { - if (ancestor_level < 0 || ancestor_level > cgrp->level) + if (ancestor_level < 0 || ancestor_level > cgroup_level(cgrp)) return NULL; return cgrp->ancestors[ancestor_level]; } diff --git a/include/trace/events/cgroup.h b/include/trace/events/cgroup.h index ba9229af9a343..0a1bc91754b5e 100644 --- a/include/trace/events/cgroup.h +++ b/include/trace/events/cgroup.h @@ -67,7 +67,7 @@ DECLARE_EVENT_CLASS(cgroup, TP_fast_assign( __entry->root = cgrp->root->hierarchy_id; __entry->id = cgroup_id(cgrp); - __entry->level = cgrp->level; + __entry->level = cgroup_level(cgrp); __assign_str(path); ), @@ -136,7 +136,7 @@ DECLARE_EVENT_CLASS(cgroup_migrate, TP_fast_assign( __entry->dst_root = dst_cgrp->root->hierarchy_id; __entry->dst_id = cgroup_id(dst_cgrp); - __entry->dst_level = dst_cgrp->level; + __entry->dst_level = cgroup_level(dst_cgrp); __assign_str(dst_path); __entry->pid = task->pid; __assign_str(comm); @@ -180,7 +180,7 @@ DECLARE_EVENT_CLASS(cgroup_event, TP_fast_assign( __entry->root = cgrp->root->hierarchy_id; __entry->id = cgroup_id(cgrp); - __entry->level = cgrp->level; + __entry->level = cgroup_level(cgrp); __assign_str(path); __entry->val = val; ), @@ -221,7 +221,7 @@ DECLARE_EVENT_CLASS(cgroup_rstat, TP_fast_assign( __entry->root = cgrp->root->hierarchy_id; __entry->id = cgroup_id(cgrp); - __entry->level = cgrp->level; + __entry->level = cgroup_level(cgrp); __entry->cpu = cpu; __entry->contended = contended; ), diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c index db72b96f9c8c8..b825f6e0a1c29 100644 --- a/kernel/bpf/helpers.c +++ b/kernel/bpf/helpers.c @@ -2577,7 +2577,7 @@ __bpf_kfunc struct cgroup *bpf_cgroup_ancestor(struct cgroup *cgrp, int level) { struct cgroup *ancestor; - if (level > cgrp->level || level < 0) + if (level > cgroup_level(cgrp) || level < 0) return NULL; /* cgrp's refcnt could be 0 here, but ancestors can still be accessed */ diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 554a02ee298ba..e011f1dd6d87f 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -5843,7 +5843,7 @@ static struct cgroup *cgroup_create(struct cgroup *parent, const char *name, struct cgroup_root *root = parent->root; struct cgroup *cgrp, *tcgrp; struct kernfs_node *kn; - int i, level = parent->level + 1; + int i, level = cgroup_level(parent) + 1; int ret; /* allocate the cgroup and its ID, 0 is reserved for the root */ @@ -5884,7 +5884,7 @@ static struct cgroup *cgroup_create(struct cgroup *parent, const char *name, goto out_stat_exit; for (tcgrp = cgrp; tcgrp; tcgrp = cgroup_parent(tcgrp)) - cgrp->ancestors[tcgrp->level] = tcgrp; + cgrp->ancestors[cgroup_level(tcgrp)] = tcgrp; /* * New cgroup inherits effective freeze counter, and diff --git a/net/netfilter/nft_socket.c b/net/netfilter/nft_socket.c index 36affbb697c2f..a5b0340924efb 100644 --- a/net/netfilter/nft_socket.c +++ b/net/netfilter/nft_socket.c @@ -64,7 +64,7 @@ static noinline int nft_socket_cgroup_subtree_level(void) if (IS_ERR(cgrp)) return PTR_ERR(cgrp); - level = cgrp->level; + level = cgroup_level(cgrp); cgroup_put(cgrp); -- 2.52.0 ^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH 2/4] cgroup: Introduce cgroup_level() helper 2025-12-17 16:27 ` [PATCH 2/4] cgroup: Introduce cgroup_level() helper Michal Koutný @ 2025-12-17 16:46 ` bot+bpf-ci 2025-12-20 14:59 ` kernel test robot 2025-12-20 21:49 ` kernel test robot 2 siblings, 0 replies; 5+ messages in thread From: bot+bpf-ci @ 2025-12-17 16:46 UTC (permalink / raw) To: mkoutny, linux-block, linux-kernel, cgroups, linux-trace-kernel, bpf, netfilter-devel, coreteam, netdev Cc: mkoutny, yukuai, axboe, tj, josef, hannes, rostedt, mhiramat, mathieu.desnoyers, ast, daniel, andrii, martin.lau, eddyz87, song, yonghong.song, john.fastabend, kpsingh, sdf, haoluo, jolsa, pablo, kadlec, fw, phil, davem, edumazet, kuba, pabeni, horms, ast, andrii, daniel, martin.lau, eddyz87, yonghong.song, clm, ihor.solodrai [-- Attachment #1: Type: text/plain, Size: 1062 bytes --] > diff --git a/block/blk-iocost.c b/block/blk-iocost.c > index a0416927d33d..b4eebe61dca7 100644 > --- a/block/blk-iocost.c > +++ b/block/blk-iocost.c [ ... ] > @@ -3003,7 +3003,7 @@ static void ioc_pd_init(struct blkg_policy_data *pd) > init_waitqueue_head(&iocg->waitq); > hrtimer_setup(&iocg->waitq_timer, iocg_waitq_timer_fn, CLOCK_MONOTONIC, HRTIMER_MODE_ABS); > > - iocg->level = blkg->blkcg->css.cgroup->level; > + iocg->level = cgroup_level(blkg->blkcg->css.cgroup) ^ Is this missing a semicolon? The line should end with a semicolon to make this valid C syntax. This will prevent the kernel from building in the ioc_pd_init() function. > > for (tblkg = blkg; tblkg; tblkg = tblkg->parent) { > struct ioc_gq *tiocg = blkg_to_iocg(tblkg); --- AI reviewed your patch. Please fix the bug or email reply why it's not a bug. See: https://github.com/kernel-patches/vmtest/blob/master/ci/claude/README.md CI run summary: https://github.com/kernel-patches/bpf/actions/runs/20310204023 ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH 2/4] cgroup: Introduce cgroup_level() helper 2025-12-17 16:27 ` [PATCH 2/4] cgroup: Introduce cgroup_level() helper Michal Koutný 2025-12-17 16:46 ` bot+bpf-ci @ 2025-12-20 14:59 ` kernel test robot 2025-12-20 21:49 ` kernel test robot 2 siblings, 0 replies; 5+ messages in thread From: kernel test robot @ 2025-12-20 14:59 UTC (permalink / raw) To: Michal Koutný, linux-block, linux-kernel, cgroups, linux-trace-kernel, bpf, netfilter-devel, coreteam, netdev Cc: oe-kbuild-all, Michal Koutný, Yu Kuai, Jens Axboe, Tejun Heo, Josef Bacik, Johannes Weiner, Steven Rostedt, Masami Hiramatsu, Mathieu Desnoyers, Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, Martin KaFai Lau, Eduard Zingerman, Song Liu, Yonghong Song, John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa, Pablo Neira Ayuso Hi Michal, kernel test robot noticed the following build errors: [auto build test ERROR on 8f0b4cce4481fb22653697cced8d0d04027cb1e8] url: https://github.com/intel-lab-lkp/linux/commits/Michal-Koutn/cgroup-Eliminate-cgrp_ancestor_storage-in-cgroup_root/20251218-004346 base: 8f0b4cce4481fb22653697cced8d0d04027cb1e8 patch link: https://lore.kernel.org/r/20251217162744.352391-3-mkoutny%40suse.com patch subject: [PATCH 2/4] cgroup: Introduce cgroup_level() helper config: sparc64-randconfig-r134-20251218 (https://download.01.org/0day-ci/archive/20251220/202512202230.1uoB5chV-lkp@intel.com/config) compiler: clang version 20.1.8 (https://github.com/llvm/llvm-project 87f0227cb60147a26a1eeb4fb06e3b505e9c7261) reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251220/202512202230.1uoB5chV-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@intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202512202230.1uoB5chV-lkp@intel.com/ All errors (new ones prefixed by >>): >> block/blk-iocost.c:3006:53: error: expected ';' after expression 3006 | iocg->level = cgroup_level(blkg->blkcg->css.cgroup) | ^ | ; 1 error generated. vim +3006 block/blk-iocost.c 2981 2982 static void ioc_pd_init(struct blkg_policy_data *pd) 2983 { 2984 struct ioc_gq *iocg = pd_to_iocg(pd); 2985 struct blkcg_gq *blkg = pd_to_blkg(&iocg->pd); 2986 struct ioc *ioc = q_to_ioc(blkg->q); 2987 struct ioc_now now; 2988 struct blkcg_gq *tblkg; 2989 unsigned long flags; 2990 2991 ioc_now(ioc, &now); 2992 2993 iocg->ioc = ioc; 2994 atomic64_set(&iocg->vtime, now.vnow); 2995 atomic64_set(&iocg->done_vtime, now.vnow); 2996 atomic64_set(&iocg->active_period, atomic64_read(&ioc->cur_period)); 2997 INIT_LIST_HEAD(&iocg->active_list); 2998 INIT_LIST_HEAD(&iocg->walk_list); 2999 INIT_LIST_HEAD(&iocg->surplus_list); 3000 iocg->hweight_active = WEIGHT_ONE; 3001 iocg->hweight_inuse = WEIGHT_ONE; 3002 3003 init_waitqueue_head(&iocg->waitq); 3004 hrtimer_setup(&iocg->waitq_timer, iocg_waitq_timer_fn, CLOCK_MONOTONIC, HRTIMER_MODE_ABS); 3005 > 3006 iocg->level = cgroup_level(blkg->blkcg->css.cgroup) 3007 3008 for (tblkg = blkg; tblkg; tblkg = tblkg->parent) { 3009 struct ioc_gq *tiocg = blkg_to_iocg(tblkg); 3010 iocg->ancestors[tiocg->level] = tiocg; 3011 } 3012 3013 spin_lock_irqsave(&ioc->lock, flags); 3014 weight_updated(iocg, &now); 3015 spin_unlock_irqrestore(&ioc->lock, flags); 3016 } 3017 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH 2/4] cgroup: Introduce cgroup_level() helper 2025-12-17 16:27 ` [PATCH 2/4] cgroup: Introduce cgroup_level() helper Michal Koutný 2025-12-17 16:46 ` bot+bpf-ci 2025-12-20 14:59 ` kernel test robot @ 2025-12-20 21:49 ` kernel test robot 2 siblings, 0 replies; 5+ messages in thread From: kernel test robot @ 2025-12-20 21:49 UTC (permalink / raw) To: Michal Koutný, linux-block, linux-kernel, cgroups, linux-trace-kernel, bpf, netfilter-devel, coreteam, netdev Cc: oe-kbuild-all, Michal Koutný, Yu Kuai, Jens Axboe, Tejun Heo, Josef Bacik, Johannes Weiner, Steven Rostedt, Masami Hiramatsu, Mathieu Desnoyers, Alexei Starovoitov, Daniel Borkmann, Andrii Nakryiko, Martin KaFai Lau, Eduard Zingerman, Song Liu, Yonghong Song, John Fastabend, KP Singh, Stanislav Fomichev, Hao Luo, Jiri Olsa, Pablo Neira Ayuso Hi Michal, kernel test robot noticed the following build errors: [auto build test ERROR on 8f0b4cce4481fb22653697cced8d0d04027cb1e8] url: https://github.com/intel-lab-lkp/linux/commits/Michal-Koutn/cgroup-Eliminate-cgrp_ancestor_storage-in-cgroup_root/20251218-004346 base: 8f0b4cce4481fb22653697cced8d0d04027cb1e8 patch link: https://lore.kernel.org/r/20251217162744.352391-3-mkoutny%40suse.com patch subject: [PATCH 2/4] cgroup: Introduce cgroup_level() helper config: nios2-allmodconfig (https://download.01.org/0day-ci/archive/20251221/202512210532.ziNaxDJf-lkp@intel.com/config) compiler: nios2-linux-gcc (GCC) 11.5.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251221/202512210532.ziNaxDJf-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@intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202512210532.ziNaxDJf-lkp@intel.com/ All error/warnings (new ones prefixed by >>): block/blk-iocost.c: In function 'ioc_pd_init': >> block/blk-iocost.c:3006:60: error: expected ';' before 'for' 3006 | iocg->level = cgroup_level(blkg->blkcg->css.cgroup) | ^ | ; 3007 | 3008 | for (tblkg = blkg; tblkg; tblkg = tblkg->parent) { | ~~~ >> block/blk-iocost.c:2988:26: warning: unused variable 'tblkg' [-Wunused-variable] 2988 | struct blkcg_gq *tblkg; | ^~~~~ vim +3006 block/blk-iocost.c 2981 2982 static void ioc_pd_init(struct blkg_policy_data *pd) 2983 { 2984 struct ioc_gq *iocg = pd_to_iocg(pd); 2985 struct blkcg_gq *blkg = pd_to_blkg(&iocg->pd); 2986 struct ioc *ioc = q_to_ioc(blkg->q); 2987 struct ioc_now now; > 2988 struct blkcg_gq *tblkg; 2989 unsigned long flags; 2990 2991 ioc_now(ioc, &now); 2992 2993 iocg->ioc = ioc; 2994 atomic64_set(&iocg->vtime, now.vnow); 2995 atomic64_set(&iocg->done_vtime, now.vnow); 2996 atomic64_set(&iocg->active_period, atomic64_read(&ioc->cur_period)); 2997 INIT_LIST_HEAD(&iocg->active_list); 2998 INIT_LIST_HEAD(&iocg->walk_list); 2999 INIT_LIST_HEAD(&iocg->surplus_list); 3000 iocg->hweight_active = WEIGHT_ONE; 3001 iocg->hweight_inuse = WEIGHT_ONE; 3002 3003 init_waitqueue_head(&iocg->waitq); 3004 hrtimer_setup(&iocg->waitq_timer, iocg_waitq_timer_fn, CLOCK_MONOTONIC, HRTIMER_MODE_ABS); 3005 > 3006 iocg->level = cgroup_level(blkg->blkcg->css.cgroup) 3007 3008 for (tblkg = blkg; tblkg; tblkg = tblkg->parent) { 3009 struct ioc_gq *tiocg = blkg_to_iocg(tblkg); 3010 iocg->ancestors[tiocg->level] = tiocg; 3011 } 3012 3013 spin_lock_irqsave(&ioc->lock, flags); 3014 weight_updated(iocg, &now); 3015 spin_unlock_irqrestore(&ioc->lock, flags); 3016 } 3017 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki ^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2025-12-20 21:49 UTC | newest] Thread overview: 5+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2025-12-17 16:27 [PATCH 0/4] Use __counted_by for ancestor arrays Michal Koutný 2025-12-17 16:27 ` [PATCH 2/4] cgroup: Introduce cgroup_level() helper Michal Koutný 2025-12-17 16:46 ` bot+bpf-ci 2025-12-20 14:59 ` kernel test robot 2025-12-20 21:49 ` kernel test robot
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).