From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mga02.intel.com ([134.134.136.20]:43779 "EHLO mga02.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932210Ab1DLOmh (ORCPT ); Tue, 12 Apr 2011 10:42:37 -0400 Subject: Re: [PATCH 10/16] iwlagn: remove un-needed configuration From: wwguy To: Stanislaw Gruszka Cc: "linville@tuxdriver.com" , "linux-wireless@vger.kernel.org" , "ipw3945-devel@lists.sourceforge.net" In-Reply-To: <20110412103905.GB2804@redhat.com> References: <1302276636-12802-1-git-send-email-wey-yi.w.guy@intel.com> <1302276636-12802-11-git-send-email-wey-yi.w.guy@intel.com> <20110412103905.GB2804@redhat.com> Content-Type: text/plain; charset="UTF-8" Date: Tue, 12 Apr 2011 07:39:55 -0700 Message-ID: <1302619195.22178.1.camel@wwguy-ubuntu> Mime-Version: 1.0 Sender: linux-wireless-owner@vger.kernel.org List-ID: On Tue, 2011-04-12 at 03:39 -0700, Stanislaw Gruszka wrote: > On Fri, Apr 08, 2011 at 08:30:30AM -0700, Wey-Yi Guy wrote: > > + lctl = iwl_pcie_link_ctl(priv); > > + if ((lctl & PCI_CFG_LINK_CTRL_VAL_L1_EN) == > > + PCI_CFG_LINK_CTRL_VAL_L1_EN) { > > + /* L1-ASPM enabled; disable(!) L0S */ > > + iwl_set_bit(priv, CSR_GIO_REG, > > + CSR_GIO_REG_VAL_L0S_ENABLED); > > + IWL_DEBUG_POWER(priv, "L1 Enabled; Disabling L0S\n"); > > + } else { > > + /* L1-ASPM disabled; enable(!) L0S */ > > + iwl_clear_bit(priv, CSR_GIO_REG, > > + CSR_GIO_REG_VAL_L0S_ENABLED); > > + IWL_DEBUG_POWER(priv, "L1 Disabled; Enabling L0S\n"); > > } > > In > > commit 1a7123cdd9f49cf1c908fb2c16d26f279c88d8c9 > Author: John W. Linville > Date: Thu Aug 5 14:39:31 2010 -0400 > > iwlwifi: disable aspm by default > > we claim to disable ASPM (including L0S), but here we > enable L0S anyway. Perhaps this is the reason why we > still get reports about device disappear: > https://bugzilla.redhat.com/show_bug.cgi?id=642966 > > We also have code from: > > commit e7cad69cdab4d6f0caadbcdd58b54214243ba98a > Author: Grumbach, Emmanuel > Date: Thu Nov 18 03:47:38 2010 -0800 > > iwlagn: Enable PCI L1 ACTIVE state after uCode has been loaded > > that seems to enable L1 ASPM as well. > > I think these bits should be removed from iwlagn driver, or better > root cause of device disappear from bus should be identified and fixed > (not sure if this is iwlagn driver/firmware problem or bios or > pci driver problem, however) to allow to use ASPM. > I also like to remove this bits, but since we still not be able to root cause the PCI-e bus disappear problem, I will like to keep the code here for now. One of the issue is we just can not re-produce the problem here, any help will be greatly appreciated. Thanks Wey