From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yinghai Lu Subject: Re: [PATCH] pci: enable MSI on 8132 Date: Sun, 22 Feb 2009 14:07:11 -0800 Message-ID: <49A1CC8F.4040202@kernel.org> References: <200902221308.55053.prakash@punnoor.de> <200902221617.21621.prakash@punnoor.de> <49A1C790.1030303@kernel.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <49A1C790.1030303@kernel.org> Sender: linux-pci-owner@vger.kernel.org To: Prakash Punnoor Cc: "Eric W. Biederman" , Robert Hancock , Jesse Barnes , Andrew Morton , david@lang.hm, Matthew Wilcox , linux-kernel , linux-scsi@vger.kernel.org, DL-MPTFusionLinux@lsi.com, linux-pci@vger.kernel.org List-Id: linux-scsi@vger.kernel.org Yinghai Lu wrote: > your system have one HT chain c51->mcp51 > > BIOS enabled HT-MSI on c51, and hda ... > it seems pci quirks should not be used enable HT MSI on mcp51 again... > another explanation: 00:09.0 RAM memory: nVidia Corporation MCP51 Host Bridge (rev a2) Subsystem: ASUSTeK Computer Inc. Device 81c0 Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx- Status: Cap+ 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- SERR- TAbort- SERR- dev, "nv_msi_ht_cap_quirk didn't locate host bridge\n"); return; } pos = pci_find_ht_capability(host_bridge, HT_CAPTYPE_SLAVE); if (pos != 0) { /* Host bridge is to HT */ ht_enable_msi_mapping(dev); return; } in c51-mcp51 case, for 00:10.1, host bridge should be 00:09.0 (mcp51) instead of 00:00.0 (c51) and ht msi for 00:10.1 is already enabled, dev 00:09.0 should not be enabled HT msi again. YH