Re: buffer cache ufs changes (preliminary ffsv2 extattr support)

2012-01-17 Thread Manuel Bouyer
On Mon, Jan 16, 2012 at 10:48:37PM +, YAMAMOTO Takashi wrote: hi, On Mon, Jan 16, 2012 at 10:00:05PM +, YAMAMOTO Takashi wrote: have you considered to separate the entity being cached from vnode? What would this buy us ? the data are intimely tied to the inode, cleaning the

Re: updated patch Re: buffer cache ufs changes (preliminary ffsv2 extattr support)

2012-01-17 Thread Manuel Bouyer
On Tue, Jan 17, 2012 at 01:21:33AM +, David Holland wrote: On Mon, Jan 16, 2012 at 10:28:57PM +0100, Manuel Bouyer wrote: I consider lfs second-class citizen at this time and if forward compat if broken for the lfs module on the branch it's probably not a big deal). I don't

NetBSD on current AMD motherboards

2012-01-17 Thread Edgar Fuß
I already asked some of these questions some time ago, but got zero replies. So I'll try to widen the audience. However, I'm not subscribed to port-i386. I must enlarge the storage capacity of our file server. This means updating from SCA to SAS and fron PCI-X to PCIe, so, new hardware. The

Re: NetBSD on current AMD motherboards

2012-01-17 Thread Thor Lancelot Simon
On Tue, Jan 17, 2012 at 08:25:04PM +0100, Edgar Fu? wrote: My main concern is SAS. It looks like there's no support for any SAS-2 controller yet, only for the (SAS-1) LSI 1068. This sounds strange. What does wasabi use in their storage products? Wasabi doesn't exist. Your best bet for a

Re: NetBSD on current AMD motherboards

2012-01-17 Thread Manuel Bouyer
On Tue, Jan 17, 2012 at 03:57:28PM -0500, Thor Lancelot Simon wrote: On Tue, Jan 17, 2012 at 08:25:04PM +0100, Edgar Fu? wrote: My main concern is SAS. It looks like there's no support for any SAS-2 controller yet, only for the (SAS-1) LSI 1068. This sounds strange. What does wasabi

Re: updated patch Re: buffer cache ufs changes (preliminary ffsv2 extattr support)

2012-01-17 Thread Manuel Bouyer
On Tue, Jan 17, 2012 at 09:25:06PM +, David Holland wrote: On Tue, Jan 17, 2012 at 07:31:17PM +0100, Manuel Bouyer wrote: So, because of modules, we can't pullup new features to netbsd-6 ? How is this different from netbsd-5? Modules are not used by default on netbsd-5

Re: updated patch Re: buffer cache ufs changes (preliminary ffsv2 extattr support)

2012-01-17 Thread Mindaugas Rasiukevicius
David Holland dholland-t...@netbsd.org wrote: On Mon, Jan 16, 2012 at 10:28:57PM +0100, Manuel Bouyer wrote: I consider lfs second-class citizen at this time and if forward compat if broken for the lfs module on the branch it's probably not a big deal). I don't consider that

Wasabi (was: NetBSD on current AMD motherboards)

2012-01-17 Thread Edgar Fuß
Wasabi doesn't exist. ???

Re: NetBSD on current AMD motherboards

2012-01-17 Thread Edgar Fuß
Your best bet for a supported storage adapter might be a RAID controller even if you don't use RAID mode. Well, but the real RAID controllers (e.g. those supporting RAID5) seem to be pretty exensive. The vendors usually keep the host interface pretty stable even as the technology on the disk

re: updated patch Re: buffer cache ufs changes (preliminary ffsv2 extattr support)

2012-01-17 Thread matthew green
David Holland dholland-t...@netbsd.org wrote: On Mon, Jan 16, 2012 at 10:28:57PM +0100, Manuel Bouyer wrote: I consider lfs second-class citizen at this time and if forward compat if broken for the lfs module on the branch it's probably not a big deal). I don't consider that

Re: NetBSD on current AMD motherboards

2012-01-17 Thread Mouse
Any RAID controller where the management interface works under NetBSD? Well, it's old enough it's more in the nature of an existence proof by example, but at work there's a NetBSD machine with a 3ware Escalade 12-port SATA RAID card. There's a management program that is depressingly poorly

Non-RAID SAS adapters (was: NetBSD on current AMD motherboards)

2012-01-17 Thread Edgar Fuß
Is there SAS controllers out there that are not RAID? Well, maybe some of them have RAID0 and RAID1. I looked for one when I needed a new tape changer, and didn't find anything with a SAS interface and no RAID (so I went fiber channel ...) LSI SAS3xxx for SAS-1, LSI SAS 9xxx for SAS-2. See

Re: PUFFS and existing file that get ENOENT

2012-01-17 Thread Emmanuel Dreyfus
YAMAMOTO Takashi y...@mwd.biglobe.ne.jp wrote: it should retry from puffs_cookie2pnode in that case. Here is a patch that works around the problem (I initially had printf to check it did go through the ENOENT case and it does). I am about to commit that and pullup to netbsd-5, except if there

Re: PUFFS and existing file that get ENOENT

2012-01-17 Thread YAMAMOTO Takashi
hi, YAMAMOTO Takashi y...@mwd.biglobe.ne.jp wrote: it should retry from puffs_cookie2pnode in that case. Here is a patch that works around the problem (I initially had printf to check it did go through the ENOENT case and it does). I am about to commit that and pullup to netbsd-5, except