Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756627AbZCVVOe (ORCPT ); Sun, 22 Mar 2009 17:14:34 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756020AbZCVVNr (ORCPT ); Sun, 22 Mar 2009 17:13:47 -0400 Received: from ogre.sisk.pl ([217.79.144.158]:54604 "EHLO ogre.sisk.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755320AbZCVVNp (ORCPT ); Sun, 22 Mar 2009 17:13:45 -0400 From: "Rafael J. Wysocki" To: Jesse Barnes Subject: [RFC][PATCH 2/2] radeonfb: Avoid open coding of PCI PM operations Date: Sun, 22 Mar 2009 22:13:13 +0100 User-Agent: KMail/1.11.1 (Linux/2.6.29-rc8-tst; KDE/4.2.1; x86_64; ; ) Cc: Benjamin Herrenschmidt , Linux PCI , pm list , LKML , Linus Torvalds , Andrew Morton References: <200903210003.55161.rjw@sisk.pl> <200903222208.22434.rjw@sisk.pl> In-Reply-To: <200903222208.22434.rjw@sisk.pl> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200903222213.14186.rjw@sisk.pl> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2215 Lines: 62 From: Rafael J. Wysocki The radeonfb driver uses open coded programming of the low power state D2 into the device, because it has to handle some chips that don't follow the specification. Avoid that by making it use __pci_set_power_state() for this purpose. Signed-off-by: Rafael J. Wysocki --- drivers/video/aty/radeon_pm.c | 26 +++----------------------- 1 file changed, 3 insertions(+), 23 deletions(-) Index: linux-2.6/drivers/video/aty/radeon_pm.c =================================================================== --- linux-2.6.orig/drivers/video/aty/radeon_pm.c +++ linux-2.6/drivers/video/aty/radeon_pm.c @@ -2507,25 +2507,6 @@ static void radeon_reinitialize_QW(struc #endif /* CONFIG_PPC_OF */ -static void radeonfb_whack_power_state(struct radeonfb_info *rinfo, pci_power_t state) -{ - u16 pwr_cmd; - - for (;;) { - pci_read_config_word(rinfo->pdev, - rinfo->pm_reg+PCI_PM_CTRL, - &pwr_cmd); - if (pwr_cmd & 2) - break; - pwr_cmd = (pwr_cmd & ~PCI_PM_CTRL_STATE_MASK) | 2; - pci_write_config_word(rinfo->pdev, - rinfo->pm_reg+PCI_PM_CTRL, - pwr_cmd); - msleep(500); - } - rinfo->pdev->current_state = state; -} - static void radeon_set_suspend(struct radeonfb_info *rinfo, int suspend) { u32 tmp; @@ -2578,11 +2559,10 @@ static void radeon_set_suspend(struct ra pci_disable_device(rinfo->pdev); pci_save_state(rinfo->pdev); /* The chip seems to need us to whack the PM register - * repeatedly until it sticks. We do that -prior- to - * calling pci_set_power_state() + * repeatedly until it sticks. However, with the 500 ms delay, + * it's better to give up after 10 unsuccessful attempts. */ - radeonfb_whack_power_state(rinfo, PCI_D2); - pci_set_power_state(rinfo->pdev, PCI_D2); + __pci_set_power_state(rinfo->pdev, PCI_D2, 10, 500); } else { printk(KERN_DEBUG "radeonfb (%s): switching to D0 state...\n", pci_name(rinfo->pdev)); -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/