On Thu, Jun 08, 2006 at 08:10:40AM -0700, Dave Hansen wrote: > On Thu, 2006-06-08 at 12:46 +0200, Herbert Poetzl wrote: > > On Wed, Jun 07, 2006 at 05:10:19PM -0700, Dave Hansen wrote: > > > > > > Take the entire "if (IS_ERR(tmp))" block, and use a goto, > > > instead. > > > > > > This should not make any functional changes. > > > > AFAICR, my test tool was done before the *at funtions > > were added to the kernel. did you extend it or test > > them in addition to the normal tests? > > No, I didn't extend it. However, I believe all of the internal kernel > implementations which are plain, like sys_mknod() call into the *at() > variants. I could look at extending that test, but I think all of the > code paths are still being exercised. > > I actually audited from the bottom up, looking for callers of the > permission functions, and anywhere that checked an inode for a > read-only superblock. well, this is a good start, but not nearly sufficient to ensure that the semantics are not changed, see my other replies to that issue ... best, Herbert > -- Dave - To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
