From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:42402) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1SeeQz-0001ND-6Y for qemu-devel@nongnu.org; Tue, 12 Jun 2012 23:39:54 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1SeeQx-0006fa-5T for qemu-devel@nongnu.org; Tue, 12 Jun 2012 23:39:52 -0400 Received: from mail-pb0-f45.google.com ([209.85.160.45]:54540) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1SeeQw-0006eu-SG for qemu-devel@nongnu.org; Tue, 12 Jun 2012 23:39:51 -0400 Received: by pbbro12 with SMTP id ro12so1780167pbb.4 for ; Tue, 12 Jun 2012 20:39:48 -0700 (PDT) Message-ID: <4FD80B7E.5090205@ozlabs.ru> Date: Wed, 13 Jun 2012 13:39:42 +1000 From: Alexey Kardashevskiy MIME-Version: 1.0 References: <4FAE07C8.7000502@ozlabs.ru> <4FD7FE82.4010804@ozlabs.ru> <1339557726.9220.84.camel@pasglop> In-Reply-To: <1339557726.9220.84.camel@pasglop> Content-Type: text/plain; charset=KOI8-R Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH] qemu msi/msix: added functions to API to set up message address and data List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Benjamin Herrenschmidt Cc: kvm@vger.kernel.org, Alex Graf , qemu-devel@nongnu.org, Alex Williamson , anthony@codemonkey.ws, David Gibson On 13/06/12 13:22, Benjamin Herrenschmidt wrote: > On Wed, 2012-06-13 at 12:44 +1000, Alexey Kardashevskiy wrote: >> Ping? > > FYI. The way it normally works on pseries is that RTAS, which is the > part of the firmware that lives inside the partition alongside the OS, > performs the accesses to configure the MSI/MSI-X. > > With qemu "pseries" machine we don't have such an RTAS currently, it's > all implemented by qemu (or rather we have a tiny 5 instructions RTAS > which does a hypercall to qemu). > > So we need to be able to do the configuration from within qemu. > > Alexey: Can this be used for supporting MSI/MSI-X with emulated devices > and virtio as well ? This is exactly to support MSIX on virtio-pci devices as there were no other MSIX users on power :) VFIO came up later (and it does not require any change in spapr-msix code). > Any chance you can do patches for that too ? I am polishing descriptions now and post them a bit later. > Cheers, > Ben. > >> >> On 12/05/12 16:48, Alexey Kardashevskiy wrote: >>> Added msi_set_address_data and msix_set_address_data in order to support >>> sPAPR-specific implementation of MSI/MSIX configuration mechanism. >>> POWER guest does not initialize these fields, it expects a supervisor to >>> get them initialized. >>> >>> Required for VFIO on POWER. >>> >>> Signed-off-by: Alexey Kardashevskiy >>> --- >>> hw/msi.c | 14 ++++++++++++++ >>> hw/msi.h | 1 + >>> hw/msix.c | 10 ++++++++++ >>> hw/msix.h | 3 +++ >>> 4 files changed, 28 insertions(+), 0 deletions(-) >>> >>> diff --git a/hw/msi.c b/hw/msi.c >>> index 5d6ceb6..124878a 100644 >>> --- a/hw/msi.c >>> +++ b/hw/msi.c >>> @@ -358,3 +358,17 @@ unsigned int msi_nr_vectors_allocated(const PCIDevice *dev) >>> uint16_t flags = pci_get_word(dev->config + msi_flags_off(dev)); >>> return msi_nr_vectors(flags); >>> } >>> + >>> +void msi_set_address_data(PCIDevice *dev, uint64_t address, uint16_t data) >>> +{ >>> + uint16_t flags = pci_get_word(dev->config + msi_flags_off(dev)); >>> + bool msi64bit = flags & PCI_MSI_FLAGS_64BIT; >>> + >>> + if (msi64bit) { >>> + pci_set_quad(dev->config + msi_address_lo_off(dev), address); >>> + } else { >>> + pci_set_long(dev->config + msi_address_lo_off(dev), address); >>> + } >>> + pci_set_word(dev->config + msi_data_off(dev, msi64bit), data); >>> +} >>> + >>> diff --git a/hw/msi.h b/hw/msi.h >>> index 3040bb0..0acf434 100644 >>> --- a/hw/msi.h >>> +++ b/hw/msi.h >>> @@ -34,6 +34,7 @@ void msi_reset(PCIDevice *dev); >>> void msi_notify(PCIDevice *dev, unsigned int vector); >>> void msi_write_config(PCIDevice *dev, uint32_t addr, uint32_t val, int len); >>> unsigned int msi_nr_vectors_allocated(const PCIDevice *dev); >>> +void msi_set_address_data(PCIDevice *dev, uint64_t address, uint16_t data); >>> >>> static inline bool msi_present(const PCIDevice *dev) >>> { >>> diff --git a/hw/msix.c b/hw/msix.c >>> index 3835eaa..c57c299 100644 >>> --- a/hw/msix.c >>> +++ b/hw/msix.c >>> @@ -414,3 +414,13 @@ void msix_unuse_all_vectors(PCIDevice *dev) >>> return; >>> msix_free_irq_entries(dev); >>> } >>> + >>> +void msix_set_address_data(PCIDevice *dev, int vector, >>> + uint64_t address, uint32_t data) >>> +{ >>> + uint8_t *table_entry = dev->msix_table_page + vector * PCI_MSIX_ENTRY_SIZE; >>> + pci_set_quad(table_entry + PCI_MSIX_ENTRY_LOWER_ADDR, address); >>> + pci_set_long(table_entry + PCI_MSIX_ENTRY_DATA, data); >>> + table_entry[PCI_MSIX_ENTRY_VECTOR_CTRL] &= ~PCI_MSIX_ENTRY_CTRL_MASKBIT; >>> +} >>> + >>> diff --git a/hw/msix.h b/hw/msix.h >>> index 5aba22b..e6bb696 100644 >>> --- a/hw/msix.h >>> +++ b/hw/msix.h >>> @@ -29,4 +29,7 @@ void msix_notify(PCIDevice *dev, unsigned vector); >>> >>> void msix_reset(PCIDevice *dev); >>> >>> +void msix_set_address_data(PCIDevice *dev, int vector, >>> + uint64_t address, uint32_t data); >>> + >>> #endif -- Alexey