make /proc/PID/*maps* namespace friendly
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest2 patches 0 repliesNew
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/2 | proc/maps: replace proc_maps_private->pid with "struct inode *inode"fs/ fs/ fs/ | +5 | 0 replies | |
| 2/2 | proc/maps: make vm_is_stack() logic namespace-friendlyfs/ fs/ include/ +1 more | +51 | 0 replies |
Reviews trailers per patch
No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.
Cover letter [PATCH -mm 0/2] make /proc/PID/*maps* namespace friendly
Hello,
Lets continue the boring proc/maps cleanups.
Eric, I do agree that the tgid vs pid logic looks strange, but I didn't even try to change it. Let's at least fix the namespace problems first. This is minor, but it is shame that /proc/pid/maps still shows the pid from the global namespace. And this is preparation for the next change.
With this change the only user of proc_maps_private->task is hold_task_mempolicy/release_task_mempolicy. I believe this code is buggy and should die, but this needs the changes in mm/mempolicy.c.
After that we can kill priv->task. task_nommu.c can be updated right now, but it would be better to change it along with task_mmu.c (we have a small complication with -ESRCH).
Oleg.
fs/proc/internal.h | 2 +- fs/proc/task_mmu.c | 29 +++++++++++++++++++++++------ fs/proc/task_nommu.c | 25 ++++++++++++++++++++++--- include/linux/mm.h | 4 ++-- mm/util.c | 23 ++++++++--------------- 5 files changed, 56 insertions(+), 27 deletions(-)