|
From: Mimi Z. <zo...@li...> - 2012-06-18 12:07:36
|
On Fri, 2012-06-15 at 13:25 +0300, Kasatkin, Dmitry wrote: > actually here is a diff > > http://git.kernel.org/?p=linux/kernel/git/kasatkin/linux-digsig.git;a=blobdiff;f=security/integrity/ima/ima_audit.c;h=b16eef4cfddbcb662c6fe10ce560d308e1cf4832;hp=21e96bf188dfcc12ff3b05226f3c7d83521dbc2b;hb=da64aee677a578a2ac66f641737fdc74e9259418;hpb=8e88fb141c9596e5efc1b72168c51484875ac5c2 > > > On Fri, Jun 15, 2012 at 1:23 PM, Kasatkin, Dmitry > <dmi...@in...> wrote: > > Hello, > > > > Actually in my tree there is a patch to show full path. > > It does reverse path walk. > > > > http://git.kernel.org/?p=linux/kernel/git/kasatkin/linux-digsig.git;a=blob;f=security/integrity/ima/ima_audit.c;h=b16eef4cfddbcb662c6fe10ce560d308e1cf4832;hb=da64aee677a578a2ac66f641737fdc74e9259418 > > > > - Dmitry Thanks Dmitry. Both this version and Peter Moody's post, which uses the existing upstream audit_log_d_path() and d_path(), use the full pathname only for auditing purposes. The discussion, here, is to simplify correlating the file measurement list hashes with filenames. We probably want a function similar to audit_log_d_format(), but returns the allocated buffer instead of freeing it, which could be used for both the measurement list and auditing. thanks, Mimi |