Re: [PATCH] rcu: use try_cmpxchg in check_cpu_stall

[Joel Fernandes]

> On Feb 28, 2023, at 3:39 PM, Joel Fernandes <joel@xxxxxxxxxxxxxxxxx> wrote:
> 
> On Tue, Feb 28, 2023 at 04:51:21PM +0100, Uros Bizjak wrote:
>> Use try_cmpxchg instead of cmpxchg (*ptr, old, new) == old in
>> check_cpu_stall.  x86 CMPXCHG instruction returns success in ZF flag, so
>> this change saves a compare after cmpxchg (and related move instruction in
>> front of cmpxchg).
> 
> In my codegen, I am not seeing mov instruction before the cmp removed, how
> can that be? The rax has to be populated with a mov before cmpxchg right?
> 
> So try_cmpxchg gives: mov, cmpxchg, cmp, jne
> Where as cmpxchg gives: mov, cmpxchg, mov, jne
> 
> So yeah you got rid of compare, but I am not seeing reduction in moves.
> Either way, I think it is an improvement due to dropping cmp so:
> 
> Acked-by: Joel Fernandes (Google) <joel@xxxxxxxxxxxxxxxxx>

Ah there probably is a lesser mov to hold the value to compare against,
but maybe not always? 

Would be nice to clarify that in the changelog anyway.

Thanks,

 - Joel

> 
> thanks,
> 
> - Joel
> 
> 
>> 
>> No functional change intended.
>> 
>> Cc: "Paul E. McKenney" <paulmck@xxxxxxxxxx>
>> Cc: Frederic Weisbecker <frederic@xxxxxxxxxx>
>> Cc: Neeraj Upadhyay <quic_neeraju@xxxxxxxxxxx>
>> Cc: Josh Triplett <josh@xxxxxxxxxxxxxxxx>
>> Cc: Steven Rostedt <rostedt@xxxxxxxxxxx>
>> Cc: Mathieu Desnoyers <mathieu.desnoyers@xxxxxxxxxxxx>
>> Cc: Lai Jiangshan <jiangshanlai@xxxxxxxxx>
>> Cc: Joel Fernandes <joel@xxxxxxxxxxxxxxxxx>
>> Signed-off-by: Uros Bizjak <ubizjak@xxxxxxxxx>
>> ---
>> kernel/rcu/tree_stall.h | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>> 
>> diff --git a/kernel/rcu/tree_stall.h b/kernel/rcu/tree_stall.h
>> index b10b8349bb2a..d81c88e66b42 100644
>> --- a/kernel/rcu/tree_stall.h
>> +++ b/kernel/rcu/tree_stall.h
>> @@ -760,7 +760,7 @@ static void check_cpu_stall(struct rcu_data *rdp)
>>    jn = jiffies + ULONG_MAX / 2;
>>    if (rcu_gp_in_progress() &&
>>        (READ_ONCE(rnp->qsmask) & rdp->grpmask) &&
>> -        cmpxchg(&rcu_state.jiffies_stall, js, jn) == js) {
>> +        try_cmpxchg(&rcu_state.jiffies_stall, &js, jn)) {
>> 
>>        /*
>>         * If a virtual machine is stopped by the host it can look to
>> @@ -778,7 +778,7 @@ static void check_cpu_stall(struct rcu_data *rdp)
>> 
>>    } else if (rcu_gp_in_progress() &&
>>           ULONG_CMP_GE(j, js + RCU_STALL_RAT_DELAY) &&
>> -           cmpxchg(&rcu_state.jiffies_stall, js, jn) == js) {
>> +           try_cmpxchg(&rcu_state.jiffies_stall, &js, jn)) {
>> 
>>        /*
>>         * If a virtual machine is stopped by the host it can look to
>> -- 
>> 2.39.2
>>