On 28/02/2008, Mark Linimon wrote:
> On Thu, Feb 28, 2008 at 02:47:28PM +0000, Chris wrote:
> > Did they push ahead with release because waiting for the mia dev?

> There's a lot of things that go into deciding when to release. The
> release cycle this time was not supposed to be as long as it was.
> We kept finding showstoppers and needing to go back and fix them
> (especially for things like TCP performance and stability).
> However, sooner or later you do have to release something. There are
> bugs in every software program ever released.
> It seems that in certain cases (we're not sure exactly which) the ata
> subsystem has regressed. There's debate on the lists about exactly
> how common this is: with bugmeister hat on, I don't think it's as
> common as some folks seem to think it is (otherwise we would be overwhelmed
> with bug reports).
> There's a lot of badly-written firmware out there, and a lot of workarounds
> for it in the current ata stack. I'm not familiar with the code, but
> fixing all this is a totally non-trivial task.
> Of course it would be nice to have several committers with sufficient
> variety of boards, and time, and experience to do a lot more QA, but
> we're a volunteer project and have to go with what we have in some cases.
> I'm open to suggestions about how we can get more people involved in
> QAing ata.
> mcl

Well I have now 3 servers running freebsd 7 and one is showing the dma
errors, a friend of mine has another 2 servers they both dont have dma
errors. So the problem doesnt seem to be frequent but is a serious
problem when it does occur assuming I am affected by the problem as my
problem could be hardware related..

freebsd-stable@freebsd.org mailing list
To unsubscribe, send any mail to "freebsd-stable-unsubscribe@freebsd.org"