Messages in this thread | | | Subject | Re: [RFC V2 PATCH] kernfs: create raw version kernfs_path_len and kernfs_path | From | "Shi, Yang" <> | Date | Mon, 29 Feb 2016 10:00:53 -0800 |
| |
On 2/27/2016 3:51 AM, Tejun Heo wrote: > Hello, > > On Sat, Feb 27, 2016 at 6:45 AM, Thomas Gleixner <tglx@linutronix.de> wrote: >>> It can be, but we can print out the ino and userland can match that up >>> with path if necessary. >> >> Wouldn't be cgroup id the better choice? > > AFAIK we aren't exposing cgroup id to userland anywhere right now. > Eventually, I think the right thing to do is using the same number for > both.
Thanks for all the comments.
Since the current tracepoints print the path length too by __trace_wb_cgroup_size and __trace_wbc_cgroup_size, but we can't get the path length if we switch to group ino. So, I'm supposed I have to drop all the *_size stuff.
And, when CONFIG_CGROUP_WRITEBACK is not enabled, __trace_wb_assign_cgroup and __trace_wbc_assign_cgroup return "strcpy(buf, "/")", so to get aligned with this, I need print out the ino of "/", right? But, the ROOT_INO may vary from different filesystems.
All of them will be addressed in V4, but it may experience some delay since I have to travel this week.
Regards, Yang
> > Thanks. >
| |