diff for duplicates of <1474580016.15303.31.camel@linux.intel.com> diff --git a/a/1.txt b/N1/1.txt index dcd9672..be908ca 100644 --- a/a/1.txt +++ b/N1/1.txt @@ -1,7 +1,7 @@ -On Thu, 2016-09-22@14:43 -0600, Jens Axboe wrote: +On Thu, 2016-09-22 at 14:43 -0600, Jens Axboe wrote: > On 09/22/2016 02:11 PM, Andy Lutomirski wrote: > > -> > On Thu, Sep 22, 2016@7:23 AM, Jens Axboe <axboe@fb.com> wrote: +> > On Thu, Sep 22, 2016 at 7:23 AM, Jens Axboe <axboe@fb.com> wrote: > > > > > > > > > On 09/16/2016 12:16 PM, Andy Lutomirski wrote: @@ -9,9 +9,9 @@ On Thu, 2016-09-22@14:43 -0600, Jens Axboe wrote: > > > > > > > > Hi all- > > > > -> > > > Here's v4 of the APST patch set.??The biggest bikesheddable +> > > > Here's v4 of the APST patch set. The biggest bikesheddable > > > > thing (I -> > > > think) is the scaling factor.??I currently have it hardcoded so +> > > > think) is the scaling factor. I currently have it hardcoded so > > > > that > > > > we wait 50x the total latency before entering a power saving > > > > state. @@ -19,21 +19,21 @@ On Thu, 2016-09-22@14:43 -0600, Jens Axboe wrote: > > > > entry > > > > latency, 5ms exit latency) after 275ms and state 4 (5mW, 2ms > > > > entry -> > > > latency, 22ms exit latency) after 1200ms.??I have the default +> > > > latency, 22ms exit latency) after 1200ms. I have the default > > > > max > > > > latency set to 25ms. > > > > > > > > FWIW, in practice, the latency this introduces seems to be well > > > > under 22ms, but my benchmark is a bit silly and I might have -> > > > measured it wrong.??I certainly haven't observed a slowdown +> > > > measured it wrong. I certainly haven't observed a slowdown > > > > just > > > > using my laptop. > > > > > > > > This time around, I changed the names of parameters after Jay -> > > > Frayensee got confused by the first try.??Now they are: +> > > > Frayensee got confused by the first try. Now they are: > > > > -> > > > ?- ps_max_latency_us in sysfs: actually controls it. -> > > > ?- nvme_core.default_ps_max_latency_us: sets the default. +> > > > - ps_max_latency_us in sysfs: actually controls it. +> > > > - nvme_core.default_ps_max_latency_us: sets the default. > > > > > > > > Yeah, they're mouthfuls, but they should be clearer now. > > > @@ -43,7 +43,7 @@ On Thu, 2016-09-22@14:43 -0600, Jens Axboe wrote: > > > knob that needs to be set. It we put it somewhere generic, then > > > at least we could potentially use it in a generic fashion. > > -> > Agreed.??I'm hoping to hear back from Rafael soon about the +> > Agreed. I'm hoping to hear back from Rafael soon about the > > dev_pm_qos > > thing. > > @@ -51,8 +51,8 @@ On Thu, 2016-09-22@14:43 -0600, Jens Axboe wrote: > > > > > > Additionally, it should not be on by default. > > -> > I think I disagree with this.??Since we don't have anything like -> > laptop-mode AFAIK, I think we do want it on by default.??For the +> > I think I disagree with this. Since we don't have anything like +> > laptop-mode AFAIK, I think we do want it on by default. For the > > server workloads that want to consume more idle power for faster > > response when idle, I think the servers should be willing to make > > this @@ -76,6 +76,6 @@ I agree with Jens, makes a lot of sense to start with this feature 'off'. To 'advertise' the feature, maybe make the feature a new selection in -Kconfig? ?Example, initially make it "EXPERIMENTAL", and later when +Kconfig? Example, initially make it "EXPERIMENTAL", and later when more devices implement this feature it can be integrated more tightly into the NVMe solution and default to on. diff --git a/a/content_digest b/N1/content_digest index 7568e36..1502199 100644 --- a/a/content_digest +++ b/N1/content_digest @@ -2,15 +2,22 @@ "ref\013b0fe95-01b8-89d7-df76-efc1aa14b27a@fb.com\0" "ref\0CALCETrV6GAZFG+cjdQrJpP7wVphfeRd=+UeTQcs5-iQYxNk6XA@mail.gmail.com\0" "ref\00a2ffd51-6e0f-7db3-8135-84317d8f77bc@fb.com\0" - "From\0james_p_freyensee@linux.intel.com (J Freyensee)\0" - "Subject\0[PATCH v4 0/3] nvme power saving\0" + "From\0J Freyensee <james_p_freyensee@linux.intel.com>\0" + "Subject\0Re: [PATCH v4 0/3] nvme power saving\0" "Date\0Thu, 22 Sep 2016 14:33:36 -0700\0" + "To\0Jens Axboe <axboe@fb.com>" + " Andy Lutomirski <luto@amacapital.net>\0" + "Cc\0linux-kernel@vger.kernel.org <linux-kernel@vger.kernel.org>" + linux-nvme@lists.infradead.org + Keith Busch <keith.busch@intel.com> + Andy Lutomirski <luto@kernel.org> + " Christoph Hellwig <hch@lst.de>\0" "\00:1\0" "b\0" - "On Thu, 2016-09-22@14:43 -0600, Jens Axboe wrote:\n" + "On Thu, 2016-09-22 at 14:43 -0600, Jens Axboe wrote:\n" "> On 09/22/2016 02:11 PM, Andy Lutomirski wrote:\n" "> > \n" - "> > On Thu, Sep 22, 2016@7:23 AM, Jens Axboe <axboe@fb.com> wrote:\n" + "> > On Thu, Sep 22, 2016 at 7:23 AM, Jens Axboe <axboe@fb.com> wrote:\n" "> > > \n" "> > > \n" "> > > On 09/16/2016 12:16 PM, Andy Lutomirski wrote:\n" @@ -18,9 +25,9 @@ "> > > > \n" "> > > > Hi all-\n" "> > > > \n" - "> > > > Here's v4 of the APST patch set.??The biggest bikesheddable\n" + "> > > > Here's v4 of the APST patch set.\302\240\302\240The biggest bikesheddable\n" "> > > > thing (I\n" - "> > > > think) is the scaling factor.??I currently have it hardcoded so\n" + "> > > > think) is the scaling factor.\302\240\302\240I currently have it hardcoded so\n" "> > > > that\n" "> > > > we wait 50x the total latency before entering a power saving\n" "> > > > state.\n" @@ -28,21 +35,21 @@ "> > > > entry\n" "> > > > latency, 5ms exit latency) after 275ms and state 4 (5mW, 2ms\n" "> > > > entry\n" - "> > > > latency, 22ms exit latency) after 1200ms.??I have the default\n" + "> > > > latency, 22ms exit latency) after 1200ms.\302\240\302\240I have the default\n" "> > > > max\n" "> > > > latency set to 25ms.\n" "> > > > \n" "> > > > FWIW, in practice, the latency this introduces seems to be well\n" "> > > > under 22ms, but my benchmark is a bit silly and I might have\n" - "> > > > measured it wrong.??I certainly haven't observed a slowdown\n" + "> > > > measured it wrong.\302\240\302\240I certainly haven't observed a slowdown\n" "> > > > just\n" "> > > > using my laptop.\n" "> > > > \n" "> > > > This time around, I changed the names of parameters after Jay\n" - "> > > > Frayensee got confused by the first try.??Now they are:\n" + "> > > > Frayensee got confused by the first try.\302\240\302\240Now they are:\n" "> > > > \n" - "> > > > ?- ps_max_latency_us in sysfs: actually controls it.\n" - "> > > > ?- nvme_core.default_ps_max_latency_us: sets the default.\n" + "> > > > \302\240- ps_max_latency_us in sysfs: actually controls it.\n" + "> > > > \302\240- nvme_core.default_ps_max_latency_us: sets the default.\n" "> > > > \n" "> > > > Yeah, they're mouthfuls, but they should be clearer now.\n" "> > > \n" @@ -52,7 +59,7 @@ "> > > knob that needs to be set. It we put it somewhere generic, then\n" "> > > at least we could potentially use it in a generic fashion.\n" "> > \n" - "> > Agreed.??I'm hoping to hear back from Rafael soon about the\n" + "> > Agreed.\302\240\302\240I'm hoping to hear back from Rafael soon about the\n" "> > dev_pm_qos\n" "> > thing.\n" "> > \n" @@ -60,8 +67,8 @@ "> > > \n" "> > > Additionally, it should not be on by default.\n" "> > \n" - "> > I think I disagree with this.??Since we don't have anything like\n" - "> > laptop-mode AFAIK, I think we do want it on by default.??For the\n" + "> > I think I disagree with this.\302\240\302\240Since we don't have anything like\n" + "> > laptop-mode AFAIK, I think we do want it on by default.\302\240\302\240For the\n" "> > server workloads that want to consume more idle power for faster\n" "> > response when idle, I think the servers should be willing to make\n" "> > this\n" @@ -85,8 +92,8 @@ "'off'.\n" "\n" "To 'advertise' the feature, maybe make the feature a new selection in\n" - "Kconfig? ?Example, initially make it \"EXPERIMENTAL\", and later when\n" + "Kconfig? \302\240Example, initially make it \"EXPERIMENTAL\", and later when\n" "more devices implement this feature it can be integrated more tightly\n" into the NVMe solution and default to on. -42bc587dd21b5209689a2f9640dddc116a47c69da7f9439f41218fb23e9ea765 +3c9c2b53cc39b137a5369c35e9323ab0d856bf1529cb8f44ffca7448bf3c0740
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.