Skip to contentSkip to conversation

posix-cpu-timers: clear TICK_DEP_BIT_POSIX_TIMER on clone

PATCH Superseded Benjamin Segall 2 participants 2 messages Oct 18, 2024 – Oct 19, 2024

Subsystems (4)
MM core Scheduler exec and binfmt Nohz, dynticks support

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Benjamin Segall

author reviewer
#0 lore raw

PATCH v2 1/1 posix-cpu-timers: clear TICK_DEP_BIT_POSIX_TIMER on clone

When we clone a new thread, we do not inherit its posix_cputimers, and clear them with posix_cputimers_init. However, this does not clear the tick dependency it creates in tsk->tick_dep_mask, and the handler does not reach the code to clear the dependency if there were no timers to begin with.

Thus if a thread has a cputimer running before cloneing/forking, that hierarchy will prevent nohz_full unless they create a cputimer of their own.

Process-wide timers do not have this problem because fork does not copy signal_struct as a baseline, it creates one from scratch.

Fixes b78783000d5c ("posix-cpu-timers: Migrate to use new tick dependency mask model")
Signed-off-by Ben Segall <[email protected]>
Suggested-by Frederic Weisbecker <[email protected]>
Cc [email protected]
---
 include/linux/tick.h | 8 ++++++++
 kernel/fork.c        | 2 ++
 2 files changed, 10 insertions(+)
include/linux/tick.h +8 -0
index 72744638c5b0..99c9c5a7252a 100644
@@ -249,16 +249,23 @@ static inline void tick_dep_set_task(struct task_struct *tsk,
 				     enum tick_dep_bits bit)
 {
 	if (tick_nohz_full_enabled())
 		tick_nohz_dep_set_task(tsk, bit);
 }
+
 static inline void tick_dep_clear_task(struct task_struct *tsk,
 				       enum tick_dep_bits bit)
 {
 	if (tick_nohz_full_enabled())
 		tick_nohz_dep_clear_task(tsk, bit);
 }
+
+static inline void tick_dep_init_task(struct task_struct *tsk)
+{
+	atomic_set(&tsk->tick_dep_mask, 0);
+}
+
 static inline void tick_dep_set_signal(struct task_struct *tsk,
 				       enum tick_dep_bits bit)
 {
 	if (tick_nohz_full_enabled())
 		tick_nohz_dep_set_signal(tsk, bit);

@@ -288,10 +295,11 @@ static inline void tick_dep_set_cpu(int cpu, enum tick_dep_bits bit) { }
 static inline void tick_dep_clear_cpu(int cpu, enum tick_dep_bits bit) { }
 static inline void tick_dep_set_task(struct task_struct *tsk,
 				     enum tick_dep_bits bit) { }
 static inline void tick_dep_clear_task(struct task_struct *tsk,
 				       enum tick_dep_bits bit) { }
+static inline void tick_dep_init_task(struct task_struct *tsk) { }
 static inline void tick_dep_set_signal(struct task_struct *tsk,
 				       enum tick_dep_bits bit) { }
 static inline void tick_dep_clear_signal(struct signal_struct *signal,
 					 enum tick_dep_bits bit) { }
kernel/fork.c +2 -0
index df8e4575ff01..9adad1ad202e 100644
@@ -103,10 +103,11 @@
 #include <linux/user_events.h>
 #include <linux/iommu.h>
 #include <linux/rseq.h>
 #include <uapi/linux/pidfd.h>
 #include <linux/pidfs.h>
+#include <linux/tick.h>
 
 #include <asm/pgalloc.h>
 #include <linux/uaccess.h>
 #include <asm/mmu_context.h>
 #include <asm/cacheflush.h>

@@ -2290,10 +2291,11 @@ __latent_entropy struct task_struct *copy_process(
 
 	task_io_accounting_init(&p->ioac);
 	acct_clear_integrals(p);
 
 	posix_cputimers_init(&p->posix_cputimers);
+	tick_dep_init_task(p);
 
 	p->io_context = NULL;
 	audit_set_context(p, NULL);
 	cgroup_fork(p);
 	if (args->kthread) {
Signature · 1 line
2.47.0.rc1.288.g06298d1525-goog
#1 lore raw
Quoted 1 line · Benjamin Segall

When we clone a new thread, we do not inherit its posix_cputimers, and

We? We neither clone something nor can we inherit anything from the kernel. Please do not impersonate code and use passive voice.

Quoted 1 line

clear them with posix_cputimers_init. However, this does not clear the

Please denote functions with brackets, i.e. fun() so its obvious.

Quoted 10 lines

tick dependency it creates in tsk->tick_dep_mask, and the handler does not reach the code to clear the dependency if there were no timers to begin with.

Thus if a thread has a cputimer running before cloneing/forking, that hierarchy will prevent nohz_full unless they create a cputimer of their own.

Process-wide timers do not have this problem because fork does not copy signal_struct as a baseline, it creates one from scratch.

And the solution to this is described where in the change log?

Thanks,

tglx