On Tue, Nov 29, 2011 at 19:42, Ben Widawsky <ben at bwidawsk.net> wrote: > On Tue, Nov 29, 2011 at 01:01:31PM +0000, Chris Wilson wrote: > > On Tue, 29 Nov 2011 10:55:04 -0200, Eugeni Dodonov < > eugeni.dodonov at intel.com> wrote: > > > This should allow even more energy saving when GPU is not in use. > > > According to the testing, this state results in around 0.1 - 0.4 W > better > > > power usage. > > > > > > No issues or regressions observed so far, but additional testing is > > > certainly welcome. > > > > The docs I saw said "not implemented; do not use". Do we have it on good > > authority that this is safe and useful to enable? And doesn't it > > require programming of more transition thresholds? > > -Chris > > Yes, I think we do have to program more stuff to make this work. Perhaps > the BIOS puts in decent values for these registers though? We would have > to restore those on reset and resume I'd guess. If BIOS doesn't use > anything there, you probably aren't even entering these states. > > Also, for posterity, there are 3 rc6 states, rc6, deep rc6, and deepest > rc6. I think deepest rc6 was recommended to avoid (though I don't recall > a specific root caused issue, just some data from Jesse, from the > windows team that it didn't seem stable). And I think deepest rc6 is > also referred to as rc7 sometimes. > We already setup the variables for both deep and deepest rc6 in our driver (GEN6_RC6p_* and GEN6_RC6pp_*), but we weren't using this additional state previously - if I understood the documentation and the code correctly, we do enable plain rc6 and deep rc6 currently. I haven't found any indications which would tell to avoid it in the latest docs, and I also haven't seen any regressions or issues with it being enabled on any of the machines, so I thought it would be worth trying that additional state as well.
