On Sunday 17 January 2016 22:09:26 Deepa Dinamani wrote:
Based on the discussion, here is how I propose to proceed:
- Series for timestamp range check and clamping
- Bug fixing patches like change all CURRENT_TIME use cases to
current_fs_time() 3. Patches for vfs to use timespec64 internally (maybe a series, if required) 4. Patches that change all fs that use vfs APIs using timestamp arguments (not a series) 5. Change individual fs to use timespec64 (not a series) 6. Change back whatever time conversion APIs left in vfs or individual fs (maybe a series, if required)
So, I don't see a need for submitting another series as all the changes now are handled on a case by case basis and no longer have a generic theme.
If everyone's in sync then I can proceed with the above plan.
Sounds good to me. Step 3 of course is the hard one, and you may run into further problems with it, as we both have in our previous attempts to crack this nut, but with step 2 before it that may become manageable.