On Tue, Jun 16, 2009 at 7:51 AM, Heinz Mauelshagen<heinzm@xxxxxxxxxx> wrote: > On Tue, 2009-06-16 at 10:09 -0400, Christoph Hellwig wrote: >> On Mon, Jun 15, 2009 at 07:21:05PM +0200, heinzm@xxxxxxxxxx wrote: >> > Hi, >> > >> > this patch series introduces the raid45 target. >> > Please include upstream. >> >> NACK, no need for another raid target, especially if it happens to be >> as broken as the raid1 one. > > You didn't review it it seems ? Will you be allowing time for review? This seems a bit late for 2.6.31, as your "Please include upstream" implies. >> Please help on making the md raid4/5 code >> more useful for your purposes (whatever that may be). > > Supporting various ATARAID raid5 mappings via dmraid long before md got > external metadata suport added. This is an opportunity to investigate what technical hurdles remain to allow dmraid to use md infrastructure [1] (or as Neil proposed [2] a new unified virtual block device infrastructure) for activating external metadata raid5 arrays. The libata/ide situation shows that there is precedence for duplicate functionality in the kernel, but my personal opinion is that we exhaust the code reuse discussion before heading down that path. Regards, Dan [1] http://marc.info/?l=linux-raid&m=123300614013042&w=2 [2] http://marc.info/?l=linux-fsdevel&m=124450400028660&w=2 -- dm-devel mailing list dm-devel@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/dm-devel
