Messages in this thread | | | Subject | [RFC PATCH 0/8] v3 contained usermode helper execution | From | Ian Kent <> | Date | Thu, 05 Feb 2015 10:33:51 +0800 |
| |
There haven't been any comments about the previous series not being an acceptable approach. Perhaps people were away, didn't notice or didn't have time.
So here's another chance to speak up.
In summary it's assumed that, since the usermode helper uses the root init namespace for process creation, using the init namespace of a container is eqivalent and sufficient when execution within a container is needed.
Thinking further about callers I believe there are cases that won't be handled properly so I've tried to work out what the current use cases are and added patches that demonstrate simple minded usage. I'm not sure at all that this is sufficient so I need feedback.
I've changed the execution to pin the calling task for the duration of the call as recommended by Jeff Layton but other than that not a lot has changed in the call back code.
It's also not clear if the request key infrastructure will continue to use a usermode callback so we'll need to wait on that.
---
Ian Kent (8): nsproxy - refactor setns() kmod - rename call_usermodehelper() flags parameter kmod - teach call_usermodehelper() to use a namespace KEYS - rename call_usermodehelper_keys() flags parameter KEYS: exec request-key within the requesting task's init namespace nfsd - use namespace if not executing in init namespace nfs - cache_lib use namespace if not executing in init namespace nfs - objlayout use namespace if not executing in init namespace
fs/nfs/cache_lib.c | 6 ++ fs/nfs/objlayout/objlayout.c | 7 ++ fs/nfsd/netns.h | 2 + fs/nfsd/nfs4recover.c | 48 ++++++++++----- include/linux/kmod.h | 20 ++++++ include/linux/nsproxy.h | 1 kernel/kmod.c | 131 ++++++++++++++++++++++++++++++++++++++---- kernel/nsproxy.c | 21 ++++--- security/keys/request_key.c | 64 +++++++++++++++++---- 9 files changed, 252 insertions(+), 48 deletions(-)
-- Ian
| |