Skip to contentSkip to conversation

tick/nohz: unexport __init-annotated tick_nohz_full_setup()

PATCH Applied Masahiro Yamada 3 participants 3 messages Jun 27, 2022 Nohz, dynticks support

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 tick/nohz: unexport __init-annotated tick_nohz_full_setup()

EXPORT_SYMBOL and __init is a bad combination because the .init.text section is freed up after the initialization. Hence, modules cannot use symbols annotated __init. The access to a freed symbol may end up with kernel panic.

modpost used to detect it, but it had been broken for a decade.

Commit 28438794aba4 ("modpost: fix section mismatch check for exported init/exit sections") fixed it so modpost started to warn it again, then this showed up:

MODPOST vmlinux.symvers
WARNING: modpost: vmlinux.o(___ksymtab_gpl+tick_nohz_full_setup+0x0): Section mismatch in reference from the variable __ksymtab_tick_nohz_full_setup to the function .init.text:tick_nohz_full_setup() The symbol tick_nohz_full_setup is exported and annotated __init
Fix this by removing the __init annotation of tick_nohz_full_setup or drop the export.

Drop the export because tick_nohz_full_setup() is only called from the built-in code in kernel/sched/isolation.c.

Fixes ae9e557b5be2 ("time: Export tick start/stop functions for rcutorture")

---

 kernel/time/tick-sched.c | 1 -
 1 file changed, 1 deletion(-)
kernel/time/tick-sched.c +0 -1
index 58a11f859ac7..30049580cd62 100644
@@ -526,7 +526,6 @@ void __init tick_nohz_full_setup(cpumask_var_t cpumask)
 	cpumask_copy(tick_nohz_full_mask, cpumask);
 	tick_nohz_full_running = true;
 }
-EXPORT_SYMBOL_GPL(tick_nohz_full_setup);
 
 static int tick_nohz_cpu_down(unsigned int cpu)
 {
Signature · 1 line
2.32.0
Tested-by on v1 1/1
#1 lore raw
Quoted 16 lines · Masahiro Yamada

EXPORT_SYMBOL and __init is a bad combination because the .init.text section is freed up after the initialization. Hence, modules cannot use symbols annotated __init. The access to a freed symbol may end up with kernel panic.

modpost used to detect it, but it had been broken for a decade.

Commit 28438794aba4 ("modpost: fix section mismatch check for exported init/exit sections") fixed it so modpost started to warn it again, then this showed up:

MODPOST vmlinux.symvers
WARNING: modpost: vmlinux.o(___ksymtab_gpl+tick_nohz_full_setup+0x0): Section mismatch in reference from the variable __ksymtab_tick_nohz_full_setup to the function .init.text:tick_nohz_full_setup() The symbol tick_nohz_full_setup is exported and annotated __init
Fix this by removing the __init annotation of tick_nohz_full_setup or drop the export.

Quoted 6 lines

Drop the export because tick_nohz_full_setup() is only called from the built-in code in kernel/sched/isolation.c.

Fixes ae9e557b5be2 ("time: Export tick start/stop functions for rcutorture")

I went through my email and notes, and cannot find whatever it was that possessed me to add tick_nohz_full_setup() to ae9e557b5be2. The other two symbols have clear motivations. Apologies for the noise!!!

Thanx, Paul

Quoted 20 lines

---

 kernel/time/tick-sched.c | 1 -
 1 file changed, 1 deletion(-)
kernel/time/tick-sched.c +0 -1
index 58a11f859ac7..30049580cd62 100644
@@ -526,7 +526,6 @@ void __init tick_nohz_full_setup(cpumask_var_t cpumask)
 	cpumask_copy(tick_nohz_full_mask, cpumask);
 	tick_nohz_full_running = true;
 }
-EXPORT_SYMBOL_GPL(tick_nohz_full_setup);
 
 static int tick_nohz_cpu_down(unsigned int cpu)
 {
Signature · 1 line
2.32.0
#2 lore raw
In reply to Masahiro Yamada
Quoted 3 lines · Masahiro Yamada

Drop the export because tick_nohz_full_setup() is only called from the built-in code in kernel/sched/isolation.c.

Thanks, applied.

Linus