The __cpuinit type of throwaway sections might have made sense some time ago when RAM was more constrained, but now the savings do not offset the cost and complications. For example, the fix in commit 5e427ec2d0 ("x86: Fix bit corruption at CPU resume time") is a good example of the nasty type of bugs that can be created with improper use of the various __init prefixes. After a discussion on LKML[1] it was decided that cpuinit should go the way of devinit and be phased out. Once all the users are gone, we can then finally remove the macros themselves from linux/init.h. This removes all the uses of the __cpuinit macros from C files in the core kernel directories (kernel, init, lib, mm, and include) that don't really have a specific maintainer. [1] https://lkml.org/lkml/2013/5/20/589
Showing the changes to kernel/workqueue.c (1 of 27 files). Show all
Unified · Split
| @@ -4644,7 +4644,7 @@ static void restore_unbound_workers_cpumask(struct worker_pool *pool, int cpu) · 4643 unmodified lines | ||
| 4644 | 4644 | * Workqueues should be brought up before normal priority CPU notifiers. |
| 4645 | 4645 | * This will be registered high priority CPU notifier. |
| 4646 | 4646 | */ |
| 4647 | −static int __cpuinit workqueue_cpu_up_callback(struct notifier_block *nfb, | |
| 4647 | +static int workqueue_cpu_up_callback(struct notifier_block *nfb, | |
| 4648 | 4648 | unsigned long action, |
| 4649 | 4649 | void *hcpu) |
| 4650 | 4650 | { |
| @@ -4697,7 +4697,7 @@ static int __cpuinit workqueue_cpu_up_callback(struct notifier_block *nfb, · 46 unmodified lines | ||
| 4697 | 4697 | * Workqueues should be brought down after normal priority CPU notifiers. |
| 4698 | 4698 | * This will be registered as low priority CPU notifier. |
| 4699 | 4699 | */ |
| 4700 | −static int __cpuinit workqueue_cpu_down_callback(struct notifier_block *nfb, | |
| 4700 | +static int workqueue_cpu_down_callback(struct notifier_block *nfb, | |
| 4701 | 4701 | unsigned long action, |
| 4702 | 4702 | void *hcpu) |
| 4703 | 4703 | { |