From patchwork Thu Feb 4 21:53:57 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Michael S. Tsirkin" X-Patchwork-Id: 8227981 Return-Path: X-Original-To: patchwork-xen-devel@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork2.web.kernel.org (Postfix) with ESMTP id E6D4BBEEE5 for ; Thu, 4 Feb 2016 21:57:22 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id E278D20395 for ; Thu, 4 Feb 2016 21:57:21 +0000 (UTC) Received: from lists.xen.org (lists.xenproject.org [50.57.142.19]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id C3C882039C for ; Thu, 4 Feb 2016 21:57:20 +0000 (UTC) Received: from localhost ([127.0.0.1] helo=lists.xen.org) by lists.xen.org with esmtp (Exim 4.72) (envelope-from ) id 1aRRqo-0006OX-0d; Thu, 04 Feb 2016 21:54:06 +0000 Received: from mail6.bemta5.messagelabs.com ([195.245.231.135]) by lists.xen.org with esmtp (Exim 4.72) (envelope-from ) id 1aRRqm-0006OS-Bl for xen-devel@lists.xensource.com; Thu, 04 Feb 2016 21:54:04 +0000 Received: from [85.158.139.211] by server-3.bemta-5.messagelabs.com id 1E/7F-13487-B78C3B65; Thu, 04 Feb 2016 21:54:03 +0000 X-Env-Sender: mst@redhat.com X-Msg-Ref: server-15.tower-206.messagelabs.com!1454622841!12027244!1 X-Originating-IP: [209.132.183.28] X-SpamReason: No, hits=0.0 required=7.0 tests=sa_preprocessor: VHJ1c3RlZCBJUDogMjA5LjEzMi4xODMuMjggPT4gNTQwNjQ=\n X-StarScan-Received: X-StarScan-Version: 7.35.1; banners=-,-,- X-VirusChecked: Checked Received: (qmail 36385 invoked from network); 4 Feb 2016 21:54:02 -0000 Received: from mx1.redhat.com (HELO mx1.redhat.com) (209.132.183.28) by server-15.tower-206.messagelabs.com with DHE-RSA-AES256-GCM-SHA384 encrypted SMTP; 4 Feb 2016 21:54:02 -0000 Received: from int-mx14.intmail.prod.int.phx2.redhat.com (int-mx14.intmail.prod.int.phx2.redhat.com [10.5.11.27]) by mx1.redhat.com (Postfix) with ESMTPS id D769DA173F; Thu, 4 Feb 2016 21:54:00 +0000 (UTC) Received: from redhat.com (vpn1-4-163.ams2.redhat.com [10.36.4.163]) by int-mx14.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with SMTP id u14Lrww5013458; Thu, 4 Feb 2016 16:53:58 -0500 Date: Thu, 4 Feb 2016 23:53:57 +0200 From: "Michael S. Tsirkin" To: qemu-devel@nongnu.org Message-ID: <1454612376-7072-45-git-send-email-mst@redhat.com> References: <1454612376-7072-1-git-send-email-mst@redhat.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <1454612376-7072-1-git-send-email-mst@redhat.com> X-Mutt-Fcc: =sent X-Scanned-By: MIMEDefang 2.68 on 10.5.11.27 Cc: Peter Maydell , xen-devel@lists.xensource.com, Eduardo Habkost , Stefano Stabellini Subject: [Xen-devel] [PULL 44/49] fix MSI injection on Xen X-BeenThere: xen-devel@lists.xen.org X-Mailman-Version: 2.1.13 Precedence: list List-Id: Xen developer discussion List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org X-Spam-Status: No, score=-4.2 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_MED, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP From: Stefano Stabellini On Xen MSIs can be remapped into pirqs, which are a type of event channels. It's mostly for the benefit of PCI passthrough devices, to avoid the overhead of interacting with the emulated lapic. However remapping interrupts and MSIs is also supported for emulated devices, such as the e1000 and virtio-net. When an interrupt or an MSI is remapped into a pirq, masking and unmasking is done by masking and unmasking the event channel. The masking bit on the PCI config space or MSI-X table should be ignored, but it isn't at the moment. As a consequence emulated devices which use MSI or MSI-X, such as virtio-net, don't work properly (the guest doesn't receive any notifications). The mechanism was working properly when xen_apic was introduced, but I haven't narrowed down which commit in particular is causing the regression. Fix the issue by ignoring the masking bit for MSI and MSI-X which have been remapped into pirqs. Signed-off-by: Stefano Stabellini Reviewed-by: Michael S. Tsirkin Signed-off-by: Michael S. Tsirkin --- include/hw/xen/xen.h | 1 + hw/pci/msi.c | 9 ++++++++- hw/pci/msix.c | 12 ++++++++++-- hw/xen/xen_pt_msi.c | 4 +--- xen-hvm-stub.c | 5 +++++ xen-hvm.c | 9 +++++++++ 6 files changed, 34 insertions(+), 6 deletions(-) diff --git a/include/hw/xen/xen.h b/include/hw/xen/xen.h index 1b81b4b..c577354 100644 --- a/include/hw/xen/xen.h +++ b/include/hw/xen/xen.h @@ -33,6 +33,7 @@ int xen_pci_slot_get_pirq(PCIDevice *pci_dev, int irq_num); void xen_piix3_set_irq(void *opaque, int irq_num, int level); void xen_piix_pci_write_config_client(uint32_t address, uint32_t val, int len); void xen_hvm_inject_msi(uint64_t addr, uint32_t data); +int xen_is_pirq_msi(uint32_t msi_data); qemu_irq *xen_interrupt_controller_init(void); diff --git a/hw/pci/msi.c b/hw/pci/msi.c index 8efa23d..85f21b8 100644 --- a/hw/pci/msi.c +++ b/hw/pci/msi.c @@ -20,6 +20,7 @@ #include "qemu/osdep.h" #include "hw/pci/msi.h" +#include "hw/xen/xen.h" #include "qemu/range.h" /* PCI_MSI_ADDRESS_LO */ @@ -254,13 +255,19 @@ void msi_reset(PCIDevice *dev) static bool msi_is_masked(const PCIDevice *dev, unsigned int vector) { uint16_t flags = pci_get_word(dev->config + msi_flags_off(dev)); - uint32_t mask; + uint32_t mask, data; + bool msi64bit = flags & PCI_MSI_FLAGS_64BIT; assert(vector < PCI_MSI_VECTORS_MAX); if (!(flags & PCI_MSI_FLAGS_MASKBIT)) { return false; } + data = pci_get_word(dev->config + msi_data_off(dev, msi64bit)); + if (xen_is_pirq_msi(data)) { + return false; + } + mask = pci_get_long(dev->config + msi_mask_off(dev, flags & PCI_MSI_FLAGS_64BIT)); return mask & (1U << vector); diff --git a/hw/pci/msix.c b/hw/pci/msix.c index 4fea7ed..eb4ef11 100644 --- a/hw/pci/msix.c +++ b/hw/pci/msix.c @@ -19,6 +19,7 @@ #include "hw/pci/msi.h" #include "hw/pci/msix.h" #include "hw/pci/pci.h" +#include "hw/xen/xen.h" #include "qemu/range.h" #define MSIX_CAP_LENGTH 12 @@ -78,8 +79,15 @@ static void msix_clr_pending(PCIDevice *dev, int vector) static bool msix_vector_masked(PCIDevice *dev, unsigned int vector, bool fmask) { - unsigned offset = vector * PCI_MSIX_ENTRY_SIZE + PCI_MSIX_ENTRY_VECTOR_CTRL; - return fmask || dev->msix_table[offset] & PCI_MSIX_ENTRY_CTRL_MASKBIT; + unsigned offset = vector * PCI_MSIX_ENTRY_SIZE; + uint32_t *data = (uint32_t *)&dev->msix_table[offset + PCI_MSIX_ENTRY_DATA]; + /* MSIs on Xen can be remapped into pirqs. In those cases, masking + * and unmasking go through the PV evtchn path. */ + if (xen_is_pirq_msi(*data)) { + return false; + } + return fmask || dev->msix_table[offset + PCI_MSIX_ENTRY_VECTOR_CTRL] & + PCI_MSIX_ENTRY_CTRL_MASKBIT; } bool msix_is_masked(PCIDevice *dev, unsigned int vector) diff --git a/hw/xen/xen_pt_msi.c b/hw/xen/xen_pt_msi.c index 5624685..9a16f2b 100644 --- a/hw/xen/xen_pt_msi.c +++ b/hw/xen/xen_pt_msi.c @@ -115,9 +115,7 @@ static int msi_msix_setup(XenPCIPassthroughState *s, assert((!is_msix && msix_entry == 0) || is_msix); - if (gvec == 0) { - /* if gvec is 0, the guest is asking for a particular pirq that - * is passed as dest_id */ + if (xen_is_pirq_msi(data)) { *ppirq = msi_ext_dest_id(addr >> 32) | msi_dest_id(addr); if (!*ppirq) { /* this probably identifies an misconfiguration of the guest, diff --git a/xen-hvm-stub.c b/xen-hvm-stub.c index a6cb5d3..c500325 100644 --- a/xen-hvm-stub.c +++ b/xen-hvm-stub.c @@ -31,6 +31,11 @@ void xen_hvm_inject_msi(uint64_t addr, uint32_t data) { } +int xen_is_pirq_msi(uint32_t msi_data) +{ + return 0; +} + void xen_ram_alloc(ram_addr_t ram_addr, ram_addr_t size, MemoryRegion *mr, Error **errp) { diff --git a/xen-hvm.c b/xen-hvm.c index 1c9fb12..6861c51 100644 --- a/xen-hvm.c +++ b/xen-hvm.c @@ -13,6 +13,7 @@ #include "hw/pci/pci.h" #include "hw/i386/pc.h" +#include "hw/i386/apic-msidef.h" #include "hw/xen/xen_common.h" #include "hw/xen/xen_backend.h" #include "qmp-commands.h" @@ -158,6 +159,14 @@ void xen_piix_pci_write_config_client(uint32_t address, uint32_t val, int len) } } +int xen_is_pirq_msi(uint32_t msi_data) +{ + /* If vector is 0, the msi is remapped into a pirq, passed as + * dest_id. + */ + return ((msi_data & MSI_DATA_VECTOR_MASK) >> MSI_DATA_VECTOR_SHIFT) == 0; +} + void xen_hvm_inject_msi(uint64_t addr, uint32_t data) { xen_xc_hvm_inject_msi(xen_xc, xen_domid, addr, data);