From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Axboe Subject: Re: [patch 3/3] Enable Aggressive Link Power management for AHCI controllers. Date: Thu, 21 Jun 2007 15:08:32 +0200 Message-ID: <20070621130831.GN18863@kernel.dk> References: <20070611184146.448266229@intel.com> <20070620142342.404856fe.kristen.c.accardi@intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from brick.kernel.dk ([80.160.20.94]:29144 "EHLO kernel.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750950AbXFUNJX (ORCPT ); Thu, 21 Jun 2007 09:09:23 -0400 Content-Disposition: inline In-Reply-To: <20070620142342.404856fe.kristen.c.accardi@intel.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Kristen Carlson Accardi Cc: jeff@garzik.org, linux-ide@vger.kernel.org, linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org, htejun@gmail.com, arjan@linux.intel.com On Wed, Jun 20 2007, Kristen Carlson Accardi wrote: > Enable Aggressive Link Power management for AHCI controllers. > > This patch will set the correct bits to turn on Aggressive > Link Power Management (ALPM) for the ahci driver. This > will cause the controller and disk to negotiate a lower > power state for the link when there is no activity (see > the AHCI 1.x spec for details). This feature is mutually > exclusive with Hot Plug, so when ALPM is enabled, Hot Plug > is disabled. ALPM will be enabled by default, but it is > settable via the scsi host syfs interface. Possible > settings for this feature are: > > Setting Effect > ---------------------------------------------------------- > min_power ALPM is enabled, and link set to enter > lowest power state (SLUMBER) when idle > Hot plug not allowed. > > max_performance ALPM is disabled, Hot Plug is allowed > > medium_power ALPM is enabled, and link set to enter > second lowest power state (PARTIAL) when > idle. Hot plug not allowed. > > Signed-off-by: Kristen Carlson Accardi A suggestion (it comes with a patch!) - default to max_power/almp off, not min_power. For two reasons: - There's such a big performance difference between the two, you really want max_power when booting. - It's a lot better to default to no change, than default to enabling something new. diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index 841cf0a..e7a2072 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -786,8 +786,7 @@ static int ahci_disable_alpm(struct ata_port *ap) return 0; } -static int ahci_enable_alpm(struct ata_port *ap, - enum scsi_host_link_pm policy) +static int ahci_enable_alpm(struct ata_port *ap, enum scsi_host_link_pm policy) { struct ahci_host_priv *hpriv = ap->host->private_data; void __iomem *port_mmio = ahci_port_base(ap); @@ -808,19 +807,19 @@ static int ahci_enable_alpm(struct ata_port *ap, return -EINVAL; } - switch(policy) { + switch (policy) { case SHOST_MAX_PERFORMANCE: - ahci_disable_alpm(ap); - ap->pm_policy = policy; - return 0; case SHOST_NOT_AVAILABLE: - case SHOST_MIN_POWER: /* * if we came here with SHOST_NOT_AVAILABLE, * it just means this is the first time we - * have tried to enable - so try to do - * min_power + * have tried to enable - default to max performance, + * and let the user go to lower power modes on request. */ + ahci_disable_alpm(ap); + ap->pm_policy = SHOST_MAX_PERFORMANCE; + return 0; + case SHOST_MIN_POWER: ap->pm_policy = SHOST_MIN_POWER; /* configure HBA to enter SLUMBER */ -- Jens Axboe