| Message ID | 20201028104825.107302-2-colin.king@canonical.com |
|---|---|
| State | New |
| Headers | show |
| Series | Fix ftrace oops/hang on RISC-V and ARM64 | expand |
On 28.10.20 11:48, Colin King wrote: > From: Zong Li <zong.li@sifive.com> > > BugLink: https://bugs.launchpad.net/bugs/1894613 > > Some architectures assume that the stopped CPUs don't make function calls > to traceable functions when they are in the stopped state. See also commit > cb9d7fd51d9f ("watchdog: Mark watchdog touch functions as notrace"). > > Violating this assumption causes kernel crashes when switching tracer on > RISC-V. > > Mark rcu_momentary_dyntick_idle() and stop_machine_yield() notrace to > prevent this. > > Fixes: 4ecf0a43e729 ("processor: get rid of cpu_relax_yield") > Fixes: 366237e7b083 ("stop_machine: Provide RCU quiescent state in multi_cpu_stop()") > Signed-off-by: Zong Li <zong.li@sifive.com> > Signed-off-by: Thomas Gleixner <tglx@linutronix.de> > Tested-by: Atish Patra <atish.patra@wdc.com> > Tested-by: Colin Ian King <colin.king@canonical.com> > Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org> > Acked-by: Paul E. McKenney <paulmck@kernel.org> > Cc: stable@vger.kernel.org > Link: https://lore.kernel.org/r/20201021073839.43935-1-zong.li@sifive.com > (cherry picked from commit 4230e2deaa484b385aa01d598b2aea8e7f2660a6 from > https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git) Signed-off-by: Colin King <colin.king@canonical.com> Acked-by: Stefan Bader <stefan.bader@canonical.com> > > --- Your sign-off is missing. And, in theory, hirsute is the new primary task, so would have to reflect whether things are needed there. I added a groovy task but left the state of the devel task unchanged. I think it probably could be fix released even though it is a bit odd right now. The actual kernel used is the copy-forward of groovy, but the unstable would be 5.10 so the question is whether that pick came from 5.10 or later and maybe needs to go into unstable, too. -Stefan > kernel/rcu/tree.c | 2 +- > kernel/stop_machine.c | 2 +- > 2 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c > index 06895ef85d69..2a52f42f64b6 100644 > --- a/kernel/rcu/tree.c > +++ b/kernel/rcu/tree.c > @@ -409,7 +409,7 @@ bool rcu_eqs_special_set(int cpu) > * > * The caller must have disabled interrupts and must not be idle. > */ > -void rcu_momentary_dyntick_idle(void) > +notrace void rcu_momentary_dyntick_idle(void) > { > int special; > > diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c > index 865bb0228ab6..890b79cf0e7c 100644 > --- a/kernel/stop_machine.c > +++ b/kernel/stop_machine.c > @@ -178,7 +178,7 @@ static void ack_state(struct multi_stop_data *msdata) > set_state(msdata, msdata->state + 1); > } > > -void __weak stop_machine_yield(const struct cpumask *cpumask) > +notrace void __weak stop_machine_yield(const struct cpumask *cpumask) > { > cpu_relax(); > } >
On 28.10.20 11:48, Colin King wrote: > From: Zong Li <zong.li@sifive.com> > > BugLink: https://bugs.launchpad.net/bugs/1894613 > > Some architectures assume that the stopped CPUs don't make function calls > to traceable functions when they are in the stopped state. See also commit > cb9d7fd51d9f ("watchdog: Mark watchdog touch functions as notrace"). > > Violating this assumption causes kernel crashes when switching tracer on > RISC-V. > > Mark rcu_momentary_dyntick_idle() and stop_machine_yield() notrace to > prevent this. > > Fixes: 4ecf0a43e729 ("processor: get rid of cpu_relax_yield") > Fixes: 366237e7b083 ("stop_machine: Provide RCU quiescent state in multi_cpu_stop()") > Signed-off-by: Zong Li <zong.li@sifive.com> > Signed-off-by: Thomas Gleixner <tglx@linutronix.de> > Tested-by: Atish Patra <atish.patra@wdc.com> > Tested-by: Colin Ian King <colin.king@canonical.com> > Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org> > Acked-by: Paul E. McKenney <paulmck@kernel.org> > Cc: stable@vger.kernel.org > Link: https://lore.kernel.org/r/20201021073839.43935-1-zong.li@sifive.com > (cherry picked from commit 4230e2deaa484b385aa01d598b2aea8e7f2660a6 from > https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git) > LGTM. With the missing s-o-b as Stefan pointed out: Acked-by: Kleber Sacilotto de Souza <kleber.souza@canonical.com> > --- > kernel/rcu/tree.c | 2 +- > kernel/stop_machine.c | 2 +- > 2 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c > index 06895ef85d69..2a52f42f64b6 100644 > --- a/kernel/rcu/tree.c > +++ b/kernel/rcu/tree.c > @@ -409,7 +409,7 @@ bool rcu_eqs_special_set(int cpu) > * > * The caller must have disabled interrupts and must not be idle. > */ > -void rcu_momentary_dyntick_idle(void) > +notrace void rcu_momentary_dyntick_idle(void) > { > int special; > > diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c > index 865bb0228ab6..890b79cf0e7c 100644 > --- a/kernel/stop_machine.c > +++ b/kernel/stop_machine.c > @@ -178,7 +178,7 @@ static void ack_state(struct multi_stop_data *msdata) > set_state(msdata, msdata->state + 1); > } > > -void __weak stop_machine_yield(const struct cpumask *cpumask) > +notrace void __weak stop_machine_yield(const struct cpumask *cpumask) > { > cpu_relax(); > } >
Applied to Groovy/master-next Thanks, Ian On 2020-10-28 10:48:25 , Colin King wrote: > From: Zong Li <zong.li@sifive.com> > > BugLink: https://bugs.launchpad.net/bugs/1894613 > > Some architectures assume that the stopped CPUs don't make function calls > to traceable functions when they are in the stopped state. See also commit > cb9d7fd51d9f ("watchdog: Mark watchdog touch functions as notrace"). > > Violating this assumption causes kernel crashes when switching tracer on > RISC-V. > > Mark rcu_momentary_dyntick_idle() and stop_machine_yield() notrace to > prevent this. > > Fixes: 4ecf0a43e729 ("processor: get rid of cpu_relax_yield") > Fixes: 366237e7b083 ("stop_machine: Provide RCU quiescent state in multi_cpu_stop()") > Signed-off-by: Zong Li <zong.li@sifive.com> > Signed-off-by: Thomas Gleixner <tglx@linutronix.de> > Tested-by: Atish Patra <atish.patra@wdc.com> > Tested-by: Colin Ian King <colin.king@canonical.com> > Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org> > Acked-by: Paul E. McKenney <paulmck@kernel.org> > Cc: stable@vger.kernel.org > Link: https://lore.kernel.org/r/20201021073839.43935-1-zong.li@sifive.com > (cherry picked from commit 4230e2deaa484b385aa01d598b2aea8e7f2660a6 from > https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git) > > --- > kernel/rcu/tree.c | 2 +- > kernel/stop_machine.c | 2 +- > 2 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c > index 06895ef85d69..2a52f42f64b6 100644 > --- a/kernel/rcu/tree.c > +++ b/kernel/rcu/tree.c > @@ -409,7 +409,7 @@ bool rcu_eqs_special_set(int cpu) > * > * The caller must have disabled interrupts and must not be idle. > */ > -void rcu_momentary_dyntick_idle(void) > +notrace void rcu_momentary_dyntick_idle(void) > { > int special; > > diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c > index 865bb0228ab6..890b79cf0e7c 100644 > --- a/kernel/stop_machine.c > +++ b/kernel/stop_machine.c > @@ -178,7 +178,7 @@ static void ack_state(struct multi_stop_data *msdata) > set_state(msdata, msdata->state + 1); > } > > -void __weak stop_machine_yield(const struct cpumask *cpumask) > +notrace void __weak stop_machine_yield(const struct cpumask *cpumask) > { > cpu_relax(); > } > -- > 2.27.0 > > > -- > kernel-team mailing list > kernel-team@lists.ubuntu.com > https://lists.ubuntu.com/mailman/listinfo/kernel-team
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 06895ef85d69..2a52f42f64b6 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -409,7 +409,7 @@ bool rcu_eqs_special_set(int cpu) * * The caller must have disabled interrupts and must not be idle. */ -void rcu_momentary_dyntick_idle(void) +notrace void rcu_momentary_dyntick_idle(void) { int special; diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index 865bb0228ab6..890b79cf0e7c 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -178,7 +178,7 @@ static void ack_state(struct multi_stop_data *msdata) set_state(msdata, msdata->state + 1); } -void __weak stop_machine_yield(const struct cpumask *cpumask) +notrace void __weak stop_machine_yield(const struct cpumask *cpumask) { cpu_relax(); }