Messages in this thread | | | Subject | Re: [RFC PATCH 1/2] Fix 1 untangling ima mess, part 2 with counters | From | Mimi Zohar <> | Date | Tue, 26 Jan 2010 17:01:01 -0500 |
| |
On Tue, 2010-01-26 at 13:03 +0000, Al Viro wrote: > OK, this stuff is in for-next, with the following changes: > * nfsd call has been moved to nfsd_open() > * patches reordered > * masking irrelevant bits (i.e. leaving only MAY_{READ,WRITE,EXEC}) > has been taken to ima_path_check(); all callers do it and it's safer that > way anyway. > > Please, see if it's OK with you in this form; other calls of dentry_open() > are separate story, we'll have to see which ones should and which ones > should not get ima_path_check(). Do you have any other problems with that > one? If not, it's going to migrate into for-linus and into the mainline; > this stuff is definitely 2.6.33 fodder.
Other than missing the export for ima_path_check() for nfs and the re-ordering not being bisect safe, they look good. I've updated the following two patches, from Eric's patchset, and would appreciate your applying them as well:
ima: rename ima_path_check to ima_file_check ima: rename PATH_CHECK to FILE_CHECK
Posting them separately.
Mimi
| |