From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Rafael J. Wysocki" Subject: [PATCH] ACPI suspend: Always use the 32-bit waking vector Date: Sat, 6 Sep 2008 13:13:01 +0200 Message-ID: <200809061313.02088.rjw@sisk.pl> References: <1220507476.4007.117.camel@yakui_zhao.sh.intel.com> <1220577437.4007.148.camel@yakui_zhao.sh.intel.com> <76780B19A496DC4B80439008DAD7076C01803B89F1@PDSMSX501.ccr.corp.intel.com> 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]:35898 "EHLO ogre.sisk.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752372AbYIFLIX convert rfc822-to-8bit (ORCPT ); Sat, 6 Sep 2008 07:08:23 -0400 In-Reply-To: <76780B19A496DC4B80439008DAD7076C01803B89F1@PDSMSX501.ccr.corp.intel.com> Content-Disposition: inline Sender: linux-acpi-owner@vger.kernel.org List-Id: linux-acpi@vger.kernel.org To: "Li, Shaohua" Cc: "Zhao, Yakui" , Matthew Garrett , "Zhang, Rui" , "lenb@kernel.org" , "linux-acpi@vger.kernel.org" , "andi@firstfloor.org" On Friday, 5 of September 2008, Li, Shaohua wrote: >=20 > >-----Original Message----- > >From: linux-acpi-owner@vger.kernel.org [mailto:linux-acpi- > >owner@vger.kernel.org] On Behalf Of Zhao Yakui > >Sent: Friday, September 05, 2008 9:17 AM > >To: Matthew Garrett > >Cc: Rafael J. Wysocki; Zhang, Rui; lenb@kernel.org; linux- > >acpi@vger.kernel.org; andi@firstfloor.org > >Subject: Re: [PATCH]: ACPI : Set 32bit and 64bit waking vector in FC= AS > >table > > > >On Thu, 2008-09-04 at 13:07 +0100, Matthew Garrett wrote: > >> On Thu, Sep 04, 2008 at 11:37:51AM +0200, Rafael J. Wysocki wrote: > >> > > so it seems that the BIOS sets =EF=BB=BFfacs->xfirmware_waking= _vector during > >> > > POST, but uses =EF=BB=BF=EF=BB=BFfacs->firmware_waking_vector = to get back during resume. > >> > > >> > So the BIOS is buggy, so let's add a quirk for it. > >> > >> Does the machine resume in Windows? If so, do we have any evidence= that > >> Windows has a quirks list to handle this case? If not, then I susp= ect > >> that Windows sets both and this is what everyone has tested agains= t. > >The laptop can be resumed on windows.(XP & Vista). And we don't know > >whether there exists the quirk list to handler this case on windows. > >Maybe what you said is right. > >In fact it is harmless when both 32bit and 64bit waking vector in FA= CS > >table are set. When the system is resumed, BIOS will transfer contro= l to > >the predefined waking vector. As we set the same waking vector, eith= er > >of them is OK. > > > >There exists the difference between 32bit and 64bit waking vector un= less > >the waking address is above 4GB. But in fact the waking address is b= elow > >1MB on most machines as the waking address needs to be accessed by B= IOS. > > > >So in most cases the 32bit and 64bit waking vector are the same valu= e. > >BIOS can transfer control to either of them. > There was discussion about this issue several months ago (intel's ml)= , looks > people forgot to take action after the discussion. The spec owner sai= d 64bit > vector is used in protected mode. That is if OS sets it, wakeup code = is > called in protected mode by BIOS. So the 64-bit vector shouldn't be u= sed. =20 Well, I read this part of the spec (2.0c, 3.0b) more carefully and it m= atches what you're saying. Moreover, my understanding of it is that we should actually _clear_ the 64-bit vector on systems that support it, because otherwise the BIOS is supposed to use it and call the wake-up code in p= rotected mode. The appended patch is based on this observation. Thanks, Rafael --- =46rom: Rafael J. Wysocki ACPI suspend: Always use the 32-bit waking vector According to the ACPI specification 2.0c and later, the 64-bit waking v= ector should be cleared and the 32-bit waking vector should be used, unless w= e want the wake-up code to be called by the BIOS in Protected Mode. Moreover,= some systems (for example HP dv5-1004nr) are known to fail to resume if the = 64-bit waking vector is used. Therefore, modify the code to clear the 64-bit = waking vector, for FACS version 1 or greater, and set the 32-bit one before su= spend. Signed-off-by: Rafael J. Wysocki --- drivers/acpi/hardware/hwsleep.c | 37 +++++++++++--------------------= ------ 1 file changed, 11 insertions(+), 26 deletions(-) Index: linux-2.6/drivers/acpi/hardware/hwsleep.c =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D --- linux-2.6.orig/drivers/acpi/hardware/hwsleep.c +++ linux-2.6/drivers/acpi/hardware/hwsleep.c @@ -78,19 +78,17 @@ acpi_set_firmware_waking_vector(acpi_phy return_ACPI_STATUS(status); } =20 - /* Set the vector */ + /* + * According to the ACPI specification 2.0c and later, the 64-bit + * waking vector should be cleared and the 32-bit waking vector shoul= d + * be used, unless we want the wake-up code to be called by the BIOS = in + * Protected Mode. Some systems (for example HP dv5-1004nr) are know= n + * to fail to resume if the 64-bit vector is used. + */ + if (facs->version >=3D 1) + facs->xfirmware_waking_vector =3D 0; =20 - if ((facs->length < 32) || (!(facs->xfirmware_waking_vector))) { - /* - * ACPI 1.0 FACS or short table or optional X_ field is zero - */ - facs->firmware_waking_vector =3D (u32) physical_address; - } else { - /* - * ACPI 2.0 FACS with valid X_ field - */ - facs->xfirmware_waking_vector =3D physical_address; - } + facs->firmware_waking_vector =3D (u32)physical_address; =20 return_ACPI_STATUS(AE_OK); } @@ -134,20 +132,7 @@ acpi_get_firmware_waking_vector(acpi_phy } =20 /* Get the vector */ - - if ((facs->length < 32) || (!(facs->xfirmware_waking_vector))) { - /* - * ACPI 1.0 FACS or short table or optional X_ field is zero - */ - *physical_address =3D - (acpi_physical_address) facs->firmware_waking_vector; - } else { - /* - * ACPI 2.0 FACS with valid X_ field - */ - *physical_address =3D - (acpi_physical_address) facs->xfirmware_waking_vector; - } + *physical_address =3D (acpi_physical_address)facs->firmware_waking_ve= ctor; =20 return_ACPI_STATUS(AE_OK); } -- 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