Re: [PATCH 14/25] sysfs: Don't use lookup_one_len_kern

Eric W. Biederman wrote: >> right? I don't think we need to duplicate the code here. Or is it >> needed for later multi-sb thing? > > Right. We can do that as well. In practice in working code > there is no real difference. > > There is a little extra uniformity in rolling it ourselves, but >

Re: [PATCH 14/25] sysfs: Don't use lookup_one_len_kern

Tejun Heo <[EMAIL PROTECTED]> writes: > On Tue, Aug 07, 2007 at 03:23:57PM -0600, Eric W. Biederman wrote: >> >> Upon inspection it appears that there is no looking of the >> inode mutex in lookup_one_len_kern and we aren't calling >> it with the inode mutex and that is wrong. >> >> So this patc

Re: [PATCH 14/25] sysfs: Don't use lookup_one_len_kern

On Tue, Aug 07, 2007 at 03:23:57PM -0600, Eric W. Biederman wrote: > > Upon inspection it appears that there is no looking of the > inode mutex in lookup_one_len_kern and we aren't calling > it with the inode mutex and that is wrong. > > So this patch rolls our own dcache insertion function that