Oleg Nesterov wrote:
On 07/26, Pavel Emelyanov wrote:
-struct pid *alloc_pid(void)
+struct pid *alloc_pid(struct pid_namespace *ns)
Why? We have the only caller, copy_process(), ns == task_active_pid_ns()
always.
task_active_pid_ns() by newly created task, not the current! That's why
we
Oleg Nesterov wrote:
On 07/26, Pavel Emelyanov wrote:
TODO: This is more an exploratory patch and modifies only interfaces
necessary to implement correct signal semantics in pid namespaces.
If the approach is feasible, we could consistently use 'kern_siginfo'
Oleg Nesterov wrote:
On 07/26, Pavel Emelyanov wrote:
+#define find_pid(pid) find_pid_ns(pid, init_pid_ns)
Again, I think find_pid() should use current's active ns, not
init_pid_ns. Just grep for find_pid/find_task_by_pid.
--- linux-2.6.23-rc1-mm1.orig/kernel/pid.c 2007-07-26
Oleg Nesterov wrote:
On 07/26, Pavel Emelyanov wrote:
This is a fix for Sukadev's patch that moved the alloc_pid() call from
do_fork() into copy_process().
... and this patch changes almost every line from Sukadev's patch.
It does. My bad :( I have reviewed Suka's patch badly and was sure
Oleg Nesterov wrote:
On 07/26, Pavel Emelyanov wrote:
static int proc_get_sb(struct file_system_type *fs_type,
int flags, const char *dev_name, void *data, struct vfsmount *mnt)
{
[...snip...]
+ if (!sb-s_root) {
+ sb-s_flags = flags;
+ err =
Oleg Nesterov wrote:
On 07/26, Pavel Emelyanov wrote:
int
kill_proc(pid_t pid, int sig, int priv)
{
- return kill_proc_info(sig, __si_special(priv), pid);
+ int ret;
+
+ rcu_read_lock();
+ ret = kill_pid_info(sig, __si_special(priv), find_pid(pid));
+
commit eab03ac7bd3e0da99eb9dc068772a85a5e3f3577 aka
[PATCH] Get rid of /proc/sys/proc was good commit except strace(1) compile
breakage it introduced:
system.c:1581: error: 'CTL_PROC' undeclared here (not in a function)
So, add dummy enum back.
Signed-off-by: Alexey Dobriyan [EMAIL
Groundhog Day?
Vasily Averin wrote:
[EMAIL PROTECTED] ~]# echo 1 /proc/timer_stats
[EMAIL PROTECTED] ~]# cat /proc/timer_stats
Timer Stats Version: v0.1
Sample period: 17.992 s
0 total events
[EMAIL PROTECTED] ~]# cat /proc/timer_stats
Timer Stats Version: v0.1
Sample period: -4366.278 s
Hello, Kirill.
Kirill Korotaev wrote:
Imho then OpenOVZ approach with multiple sysfs trees is better.
it allows to use cached dentries with moultiple sysfs mounts
each having different view.
It also allows to hide hw-related entries and events from the containers
and has quite little
Kirill Korotaev wrote:
Tejun Heo wrote:
I thought something like supermount plus some twists or fuse based sysfs
proxy would fit better. Dunno whether or how uevent and polling stuff
can work that way tho. Note that sysfs no longer keeps dentries and
inodes pinned. It might make the shared
Tejun Heo [EMAIL PROTECTED] writes:
Hello,
Okay, some questions.
* What do you think about not allowing duplicate names across different
tags? ie. there's only one ethX anywhere but it's visible only in a
specific namespace (and maybe in the default global one). Or does
everyone need
On 07/30, Pavel Emelyanov wrote:
Oleg Nesterov wrote:
+
+ nfree = 0;
+ for (i = 0; i PIDMAP_ENTRIES; i++)
+ nfree += atomic_read(pid_ns-pidmap[i].nr_free);
+
+ /*
+* If pidmap has entries for processes other than 0 and 1, retry.
+*/
+ if (nfree (BITS_PER_PAGE
Hi,
On 26.07.2007 10:00, Tejun Heo wrote:
Okay, some questions.
* What do you think about not allowing duplicate names across different
tags? ie. there's only one ethX anywhere but it's visible only in a
specific namespace (and maybe in the default global one). Or does
everyone need its
Pavel Emelianov [EMAIL PROTECTED] wrote:
| Oleg Nesterov wrote:
| On 07/26, Pavel Emelyanov wrote:
| This is a fix for Sukadev's patch that moved the alloc_pid() call from
| do_fork() into copy_process().
|
| ... and this patch changes almost every line from Sukadev's patch.
|
| It does. My bad
Pavel Emelianov [EMAIL PROTECTED] wrote:
| Oleg Nesterov wrote:
| On 07/26, Pavel Emelyanov wrote:
| TODO: This is more an exploratory patch and modifies only
| interfaces
| necessary to implement correct signal semantics in pid namespaces.
|
| If the approach is feasible, we
Ugh. I need to step back and carefully define what I'm seeing but it
looks like the current sysfs locking is wrong.
I'm starting to find little inconsistencies all over the place
such as:
Which lock actually protects sd-s_children?
- It isn't sysfs_mutex. (see sysfs_lookup)
- It isn't
+ lock_meta_page(page);
+ /*
+ * Check if somebody else beat us to allocating the meta_page
+ */
+ race_mp = page_get_meta_page(page);
+ if (race_mp) {
+ kfree(mp);
+ mp = race_mp;
+ atomic_inc(mp-ref_cnt);
+
Hello,
Eric W. Biederman wrote:
Ugh. I need to step back and carefully define what I'm seeing but it
looks like the current sysfs locking is wrong.
I'm starting to find little inconsistencies all over the place
such as:
Which lock actually protects sd-s_children?
- It isn't
Hello,
Eric W. Biederman wrote:
How close are we to the point where we can get mount sysfs multiple
times and get multiple dentry trees with different super blocks?
Yeah, that sounds much better. We only have to pay attention to getting
sysfs_dirent tree correct. The rest can be done by just
Tejun Heo [EMAIL PROTECTED] writes:
Hello,
Eric W. Biederman wrote:
Ugh. I need to step back and carefully define what I'm seeing but it
looks like the current sysfs locking is wrong.
I'm starting to find little inconsistencies all over the place
such as:
Which lock actually protects
Eric W. Biederman wrote:
What do we use inode-i_mutex for? I think we might be able
to kill that.
I'm starting to wonder if we can completely remove sysfs
from grabbing inode-i_mutex.
i_mutex is grabbed when dentry and inode locking requires it. It's not
used to protect sysfs internal
john stultz wrote:
On 7/29/07, Vasily Averin [EMAIL PROTECTED] wrote:
I've investigated why my testnode freezes. When I found that node is freezed
again I've started to press Sysrq keys and noticed the following negative
time jump.
Could anybody please help me to understand the reasons of
22 matches
Mail list logo
