From: Dan Smith <danms@xxxxxxxxxx> Subject: Re: dm-userspace (no in-kernel cache version) Date: Tue, 12 Sep 2006 21:28:58 -0700 > FT> Surely, we need to replace the request list with hash list. > > I have done that now. It helps performance quite a bit, although it > does not perform at the same level with lots of threads. I'll keep > working on it. I see. Thanks. > FT> Yep, I dropped some of the features because of my laziness, though > FT> if endio means that the kernel notifies user-space of I/O > FT> completion, I think that I implemented it. > > We need more than just notification of endio, but also the ability to > delay the endio completion until userspace has acknowledged the > endio. This is crucial for us to maintain metadata consistency in the > CoW case. I'm not sure how this works. Have you explained the details of this feature in the earlier thread? > FT> One possible feature is support for multiple destinations. If user > FT> space can tell kernel to write multiple devices, we can implement > FT> kinda RAID daemons in user space. > > Yes, my original version supported this, which would allow RAID from > userspace, as well as lots of other neat tricks :) I see. I don't think that you need to implement it now (simple code is always better for mainline inclusion), however, it would be nice to have dmu_event structure for this. -- dm-devel mailing list dm-devel@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/dm-devel
