Messages in this thread | | | Date | Wed, 29 Aug 2012 11:32:26 +0300 | From | Alexey Dobriyan <> | Subject | Re: [PATCH] fs/proc: Move kfree outside pde_unload_lock |
| |
On Tue, Aug 28, 2012 at 09:11:57PM -0700, Eric Dumazet wrote: > On Tue, 2012-08-28 at 23:38 +0300, Alexey Dobriyan wrote: > > > Nothing can stop RCU! > > > > After running "modprobe;rmmod" in a loop and "cat" in another loop for a while > > rmmod got stuck in D-state inside remove_proc_entry() with trace amounts of CPU time > > being consumed. > > > > It didn't oopsed, though. > > Thanks ! > > I'll polish this patch once LKS/LPC is over... > > What particular module and/or proc file did you use for your tests ?
Just dummy one.
#include <linux/init.h> #include <linux/proc_fs.h> #include <linux/seq_file.h>
static int foo_proc_show(struct seq_file *m, void *v) { seq_puts(m, "foo\n"); return 0; }
static int foo_proc_open(struct inode *inode, struct file *file) { return single_open(file, foo_proc_show, NULL); }
static const struct file_operations foo_proc_ops = { .open = foo_proc_open, .read = seq_read, .llseek = seq_lseek, .release = single_release, };
static int __init foo_module_init(void) { proc_create("foo", 0, NULL, &foo_proc_ops); return 0; }
static void __exit foo_module_exit(void) { remove_proc_entry("foo", NULL); } module_init(foo_module_init); module_exit(foo_module_exit);
| |