From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Rafael J. Wysocki" Subject: Re: [PATCH] ACPI: evaluate _PS3 when entering D3 Cold Date: Mon, 9 Apr 2012 01:41:36 +0200 Message-ID: <201204090141.36766.rjw@sisk.pl> References: <1333217910-29579-1-git-send-email-aaron.lu@amd.com> <201204011049.08609.rjw@sisk.pl> Mime-Version: 1.0 Content-Type: Text/Plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from ogre.sisk.pl ([217.79.144.158]:59158 "EHLO ogre.sisk.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756148Ab2DHXhJ convert rfc822-to-8bit (ORCPT ); Sun, 8 Apr 2012 19:37:09 -0400 In-Reply-To: Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: huang ying Cc: Zhang Rui , Lin Ming , Aaron Lu , Len Brown , linux-acpi@vger.kernel.org, linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org, Andiry Xu , Alex He , Huang Ying On Thursday, April 05, 2012, huang ying wrote: > Hi, Rafael, Hi, > On Sun, Apr 1, 2012 at 4:49 PM, Rafael J. Wysocki wrote= : > > On Sunday, April 01, 2012, Zhang Rui wrote: > >> On =E6=97=A5, 2012-04-01 at 09:23 +0200, Rafael J. Wysocki wrote: > >> > Hi, > >> > > >> > Sorry for the delayed response, I've been travelling recently. > >> > > >> > On Sunday, April 01, 2012, Lin Ming wrote: > >> > > On Sun, 2012-04-01 at 13:56 +0800, Aaron Lu wrote: > >> > > > Hi, > >> > > > > >> > > > On Sun, Apr 01, 2012 at 01:27:33PM +0800, Lin Ming wrote: > >> > > > > > - if (device->power.states[state].flags.explic= it_set) { > >> > > > > > + /* If state is D3 Cold, try to evaluate _PS3= first */ > >> > > > > > + if (state =3D=3D ACPI_STATE_D3_COLD) { > >> > > > > > + explicit_set =3D (ps - 1)->flags.exp= licit_set; > >> > > > > > + object_name[3] -=3D 1; > >> > > > > > + } > >> > > > > > >> > > > > I'm not sure whether this works or not. > >> > > > > > >> > > > > From ACPI spec, > >> > > > > > >> > > > > _PS3 "is used to put the specific device into its D3hot or= D3 state" > >> > > > > > >> > > > > D3 neither means D3hot nor D3cold. It's an old term before= D3hot and > >> > > > > D3cold were introduced. > >> > > > I guess D3 has to mean something, right? :-) > >> > > >> > Well, not necessarily. > >> > > >> > The problem is what state the _PS3 method puts the device into: = D3_hot or > >> > D3_cold. > >> > > >> > Unfortunately, as far as I can say, ACPI 4.0 didn't specify any = "official" > >> > mapping between the "old" D3 and the "new" D3_{hod|cold} states,= so we need to > >> > figure out something. In my opinion, the only reasonable approa= ch is to > >> > assume that the state _PS3 puts the device into is always D3_col= d, becuase > >> > _PS3 may remove power completely from the device. It may not do= that, but > >> > we _must_ assume it does that in general. > >> > > >> There is a problem that I can think of. > >> Say currently, ACPI always returns D3 when _PS3 exists. > > > > I'm not sure what you mean exactly, but please see below. > > > >> And this "ACPI_STATE_D3" is translated to PCI_D3hot. > >> But with this approach, we're going to put these devices to PCI_D3= cold > >> instead, right? > >> I'm not against this approach, but this may affect a lot of PCI de= vices, > >> which we need to take care of, no? > > > > Do you mean acpi_pm_device_sleep_state() should be modified? It do= esn't > > refer to whether or not _PS3 exists, as far as I can say. > > > > In the acpi_pci_set_power_state() code path, on the other hand, the= re are > > two potentially problematic situations when PCI wants to put the de= vice into > > PCI_D3hot (which need not map 1:1 onto ACPI D3_hot), one of which i= s when _PS3 > > is present, and the other is when neither _PS3, nor _PR3 is present= =2E > > > > If _PS3 is present, then _PR3 may or may not be present. In the la= tter case > > we can only execute _PS3 in the hope it does the right thing, but a= s long > > as we restore the device's configuration registers while resuming i= t (which is > > done by all of our PCI device resume callback routines as far as I = can say), > > the only possible difference is the resume latency (which may be gr= eater if > > power is removed from the device entirely). >=20 > Another difference between D3Hot and D3Cold for PCI devices is config > space availability. That is, in D3Hot, you can access D3Hot, while i= n > D3Cold you can not do that. For example, PME poll logic need to be > disabled if we put device into D3Cold. We're not talking about PCI here. PCI D3hot/D3cold is actually well de= fined, while the ACPI "couterparts" aren't. And BTW I know the properties of = the PCI power management states. :-) > > However, in that case we shouldn't > > turn off the device's power resources after _PS3 has been executed = (if we > > turned them off, power would be removed from the device, which woul= dn't be > > what PCI wanted). So, to handle this particular case we need to pa= ss > > ACPI_STATE_D3_HOT to acpi_bus_set_power(), meaning "avoid going int= o D3_cold, > > if possible". > > > > In both _PS3 and _PR3 are present, we should evaluate _PS3 and then= turn off > > the power resources listed as "off" by _PR3 (and turn on the power = resoruces > > listed by it as "on"), but we need to restore the configuration reg= isters of > > the device while resuming it. I think this is handled correctly wi= thout > > modifications. > > > > If neither _PS3 nor _PR3 is present, we shouldn't turn off the devi= ce's > > power resources, because PCI doesn't want power to be removed from = the device. >=20 > For PCI device plugged into system via slot (not integrated into PCH > or motherboard), there is no ACPI handle associate with it, so that > there are neither _PS3 nor _PR3 presented. But it is still possible > to turn off the device power via the associated PCIe port, which has > _PS3 and/or _PR3 presented. I think that situation is reasonable too= =2E Again, this hasn't anything to do with ACPI. We're discussing standard interfaces exposed by ACPI. That is, if I sa= y "to turn of the device's power resources" I mean to call _OFF for all o= f the power resources listed by _PR0 for that device. Nothing more or less t= han that. > > In summary, if PCI wants the device to be put into PCI_D3hot and _P= S3 is > > present, we should evaluate _PS3. However, we shouldn't turn the d= evice's > > power resources off unless _PR3 is present, in which case we can tu= rn off > > the power resources listed by it as "off". >=20 > How to turn the device's power resources off without _PR3? It may be > possible via PCIe port as I said above. Do you mean that? Or > something else? If _PR0 is present, it returns the list of power resources needed by th= e device. If you turn them all off, the assumption is that power has bee= n removed from the device, so it is in D3(cold). PCI hotplug (and other standards) provide alternative means for removin= g power from devices, but that's beyond the scope of what is being discus= sed in this thread. Thanks, Rafael -- To unsubscribe from this list: send the line "unsubscribe linux-acpi" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html