Re: [PATCH 21/22] HWPOISON: send uevent to report memory corruption

[Wu Fengguang <fengguang.wu@intel.com>]

On Mon, Jun 15, 2009 at 02:29:34PM +0800, Andi Kleen wrote:
> On Mon, Jun 15, 2009 at 10:45:41AM +0800, Wu Fengguang wrote:
> > This allows the user space to do some flexible policies.
> > For example, it may either do emergency sync/shutdown
> > or to schedule reboot at some convenient time, depending
> > on the severeness of the corruption.
> > 
> 
> I don't think it's a good idea to export that much detailed information.
> That would become a stable ABI, but might not be possible to keep
> all these details stable. e.g. map count or reference count are
> internal implementation details that shouldn't be exposed.
> And what is an user space application to do with the inode? Run
> find -inum? 

I had plan to export file path too, if available. But you are right,
we should really think twice before exporting them.

> Also we already report the event using low level logging mechanism.
> in a relatively stable form.
> 
> It's also unclear to me what an application would do with that much
> detail.
> 
> I would suggest to drop this part and the earlier flags move.
> 
> Please only bug fixes are this stage.

OK, I'll get rid of the uevent patches numbered 16-21 in the next release.

Thanks,
Fengguang

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>