Re: [RFC][PATCH 00/10] On inode::i_count and the usage vs reference count issue

2017-02-24 Thread David Windsor
On Fri, Feb 24, 2017 at 11:43 AM, Christoph Hellwig wrote: > Usage counts are common and useful, so for now they should stay as-is > and if people can came up with a useful primitive for them we can > consider implementing it. > While developing the refcount_t API, we used coccinelle to find area

Re: [RFC][PATCH 00/10] On inode::i_count and the usage vs reference count issue

2017-02-24 Thread Al Viro
On Fri, Feb 24, 2017 at 04:43:29PM +0100, Peter Zijlstra wrote: > There's a number of options here: > > - I'm not completely insane, and these patches can be made to work. > > - We decide usage-counts are useful and try and support them in refcount_t; >this has the down-side that people c

Re: [RFC][PATCH 00/10] On inode::i_count and the usage vs reference count issue

2017-02-24 Thread Peter Zijlstra
On Fri, Feb 24, 2017 at 08:43:30AM -0800, Christoph Hellwig wrote: > Usage counts are common and useful, so for now they should stay as-is > and if people can came up with a useful primitive for them we can > consider implementing it. > > Trying to shoe-horn everything into refcount_t is a horribl

Re: [RFC][PATCH 00/10] On inode::i_count and the usage vs reference count issue

2017-02-24 Thread Christoph Hellwig
Usage counts are common and useful, so for now they should stay as-is and if people can came up with a useful primitive for them we can consider implementing it. Trying to shoe-horn everything into refcount_t is a horrible idea.