xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

saner FASYNC handling on file close

Browse Source 
As it is, all instances of ->release() for files that have ->fasync()
need to remember to evict file from fasync lists; forgetting that
creates a hole and we actually have a bunch that *does* forget.

So let's keep our lives simple - let __fput() check FASYNC in
file->f_flags and call ->fasync() there if it's been set.  And lose that
crap in ->release() instances - leaving it there is still valid, but we
don't have to bother anymore.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This commit is contained in:
Al Viro
2008-10-31 23:28:30 +00:00
committed by Linus Torvalds
parent 3318a386e4
commit 233e70f422
36 changed files with 13 additions and 109 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
drivers/message/i2o/i2o_config.c
View File

@@ -1097,28 +1097,17 @@ static int cfg_fasync(int fd, struct file *fp, int on)
static int cfg_release(struct inode *inode, struct file *file)
{
ulong id = (ulong) file->private_data;
struct i2o_cfg_info *p1, *p2;
struct i2o_cfg_info *p, **q;
unsigned long flags;
lock_kernel();
p1 = p2 = NULL;
spin_lock_irqsave(&i2o_config_lock, flags);
for (p1 = open_files; p1;) {
if (p1->q_id == id) {
if (p1->fasync)
cfg_fasync(-1, file, 0);
if (p2)
p2->next = p1->next;
else
open_files = p1->next;
kfree(p1);
for (q = &open_files; (p = *q) != NULL; q = &p->next) {
if (p->q_id == id) {
*q = p->next;
kfree(p);
break;
}
p2 = p1;
p1 = p1->next;
}
spin_unlock_irqrestore(&i2o_config_lock, flags);
unlock_kernel();