Messages in this thread | | | Date | Fri, 18 Dec 2009 13:04:28 +0900 | From | Hiroshi Shimamoto <> | Subject | Re: revert "config FS_JOURNAL_INFO" |
| |
Alexey Dobriyan wrote: > Can we please revert commit e4c570c4cb7a95dbfafa3d016d2739bf3fdfe319 > "task_struct: make journal_info conditional"
I'm fine to revert it, because it seems merged accidentally.
> > I think I gave a good enough arguments to not merge it. > To iterate: > * patch makes impossible to start using ext3 on EXT3_FS=n kernels > without reboot. > * this is done only for one pointer on task_struct
I don't think it's only one pointer. There might be a lot of "only one pointer".
> > None of config options which define task_struct are tristate directly > or effectively.
So we never allow to make memory usage small with providing an option to remove unused area, right? Of cause, that option should be handled carefully. If I want to reduce memory usage by this way, should I keep this kind of patches out of tree?
thanks, Hiroshi
| |