locking/csd_lock: Remove per-CPU data indirection from CSD lock debugging

The diagnostics added by this commit were extremely useful in one instance:

a5aabace ("locking/csd_lock: Add more data to CSD lock debugging")

However, they have not seen much action since, and there have been some
concerns expressed that the complexity is not worth the benefit.

Therefore, manually revert the following commit preparatory commit:

de7b09ef ("locking/csd_lock: Prepare more CSD lock debugging")
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Juergen Gross <jgross@suse.com>
Link: https://lore.kernel.org/r/20230321005516.50558-3-paulmck@kernel.org

kernel/smp.c

@@ -31,12 +31,8 @@
 
 #define CSD_TYPE(_csd)	((_csd)->node.u_flags & CSD_FLAG_TYPE_MASK)
 
-struct cfd_percpu {
-	call_single_data_t	csd;
-};
-
 struct call_function_data {
-	struct cfd_percpu	__percpu *pcpu;
+	call_single_data_t	__percpu *csd;
 	cpumask_var_t		cpumask;
 	cpumask_var_t		cpumask_ipi;
 };
@@ -59,8 +55,8 @@ int smpcfd_prepare_cpu(unsigned int cpu)
 		free_cpumask_var(cfd->cpumask);
 		return -ENOMEM;
 	}
-	cfd->pcpu = alloc_percpu(struct cfd_percpu);
-	if (!cfd->pcpu) {
+	cfd->csd = alloc_percpu(call_single_data_t);
+	if (!cfd->csd) {
 		free_cpumask_var(cfd->cpumask);
 		free_cpumask_var(cfd->cpumask_ipi);
 		return -ENOMEM;
@@ -75,7 +71,7 @@ int smpcfd_dead_cpu(unsigned int cpu)
 
 	free_cpumask_var(cfd->cpumask);
 	free_cpumask_var(cfd->cpumask_ipi);
-	free_percpu(cfd->pcpu);
+	free_percpu(cfd->csd);
 	return 0;
 }
 
@@ -730,7 +726,7 @@ static void smp_call_function_many_cond(const struct cpumask *mask,
 
 		cpumask_clear(cfd->cpumask_ipi);
 		for_each_cpu(cpu, cfd->cpumask) {
-			call_single_data_t *csd = &per_cpu_ptr(cfd->pcpu, cpu)->csd;
+			call_single_data_t *csd = per_cpu_ptr(cfd->csd, cpu);
 
 			if (cond_func && !cond_func(cpu, info))
 				continue;
@@ -774,7 +770,7 @@ static void smp_call_function_many_cond(const struct cpumask *mask,
 		for_each_cpu(cpu, cfd->cpumask) {
 			call_single_data_t *csd;
 
-			csd = &per_cpu_ptr(cfd->pcpu, cpu)->csd;
+			csd = per_cpu_ptr(cfd->csd, cpu);
 			csd_lock_wait(csd);
 		}
 	}