From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:46721) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1cTjDD-00047L-Is for qemu-devel@nongnu.org; Wed, 18 Jan 2017 00:55:13 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1cTjDA-0006xw-Em for qemu-devel@nongnu.org; Wed, 18 Jan 2017 00:55:11 -0500 Received: from mx0b-001b2d01.pphosted.com ([148.163.158.5]:57684 helo=mx0a-001b2d01.pphosted.com) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1cTjDA-0006x2-8i for qemu-devel@nongnu.org; Wed, 18 Jan 2017 00:55:08 -0500 Received: from pps.filterd (m0098417.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.16.0.20/8.16.0.20) with SMTP id v0I5rVP3082954 for ; Wed, 18 Jan 2017 00:55:07 -0500 Received: from e19.ny.us.ibm.com (e19.ny.us.ibm.com [129.33.205.209]) by mx0a-001b2d01.pphosted.com with ESMTP id 281xc2fcaw-1 (version=TLSv1.2 cipher=AES256-SHA bits=256 verify=NOT) for ; Wed, 18 Jan 2017 00:55:07 -0500 Received: from localhost by e19.ny.us.ibm.com with IBM ESMTP SMTP Gateway: Authorized Use Only! Violators will be prosecuted for from ; Wed, 18 Jan 2017 00:55:06 -0500 Date: Wed, 18 Jan 2017 13:55:02 +0800 From: Dong Jia Shi References: <20170112072513.98411-1-bjsdjshi@linux.vnet.ibm.com> <20170112072513.98411-10-bjsdjshi@linux.vnet.ibm.com> <20170117155313.02de9785@t450s.home> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170117155313.02de9785@t450s.home> Message-Id: <20170118055502.GR30301@bjsdjshi@linux.vnet.ibm.com> Subject: Re: [Qemu-devel] [PATCH RFC v2 09/12] vfio/ccw: get irqs info and set the eventfd fd List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alex Williamson Cc: Dong Jia Shi , kvm@vger.kernel.org, linux-s390@vger.kernel.org, qemu-devel@nongnu.org, renxiaof@linux.vnet.ibm.com, cornelia.huck@de.ibm.com, borntraeger@de.ibm.com, agraf@suse.com, pmorel@linux.vnet.ibm.com, pasic@linux.vnet.ibm.com, wkywang@linux.vnet.ibm.com * Alex Williamson [2017-01-17 15:53:13 -0700]: > On Thu, 12 Jan 2017 08:25:10 +0100 > Dong Jia Shi wrote: > > > From: Xiao Feng Ren > > > > vfio-ccw resorts to the eventfd mechanism to communicate with userspace. > > We fetch the irqs info via the ioctl VFIO_DEVICE_GET_IRQ_INFO, > > register a event notifier to get the eventfd fd which is sent > > to kernel via the ioctl VFIO_DEVICE_SET_IRQS, then we can implement > > read operation once kernel sends the signal. > > > > Signed-off-by: Xiao Feng Ren > > --- > > hw/vfio/ccw.c | 102 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > 1 file changed, 102 insertions(+) > > > > diff --git a/hw/vfio/ccw.c b/hw/vfio/ccw.c > > index 93394c2..c6bfce7 100644 > > --- a/hw/vfio/ccw.c > > +++ b/hw/vfio/ccw.c > > @@ -21,6 +21,7 @@ > > #include "hw/vfio/vfio-common.h" > > #include "hw/s390x/s390-ccw.h" > > #include "hw/s390x/ccw-device.h" > > +#include "qemu/error-report.h" > > #include "standard-headers/asm-s390/vfio_ccw.h" > > > > #define TYPE_VFIO_CCW "vfio-ccw" > > @@ -30,6 +31,7 @@ typedef struct VFIOCCWDevice { > > uint64_t io_region_size; > > uint64_t io_region_offset; > > struct ccw_io_region *io_region; > > + EventNotifier io_notifier; > > } VFIOCCWDevice; > > > > static void vfio_ccw_compute_needs_reset(VFIODevice *vdev) > > @@ -54,6 +56,98 @@ static void vfio_ccw_reset(DeviceState *dev) > > ioctl(vcdev->vdev.fd, VFIO_DEVICE_RESET); > > } > > > > +static void vfio_ccw_io_notifier_handler(void *opaque) > > +{ > > + VFIOCCWDevice *vcdev = opaque; > > + > > + if (!event_notifier_test_and_clear(&vcdev->io_notifier)) { > > + return; > > + } > > +} > > + > > +static void vfio_ccw_register_io_notifier(VFIOCCWDevice *vcdev, Error **errp) > > +{ > > + VFIODevice *vdev = &vcdev->vdev; > > + struct vfio_irq_info *irq_info; > > + struct vfio_irq_set *irq_set; > > + size_t argsz; > > + int32_t *pfd; > > + > > + if (vdev->num_irqs != VFIO_CCW_NUM_IRQS) { > > + error_setg(errp, "vfio: unexpected number of io irqs %u", > > + vdev->num_irqs); > > + return; > > + } > > Same issue here as with region info, we may find reasons to add new > interrupts in the kernel, it's happened for PCI support with things > like device error recovery and requests to release the device. Don't > lock us into never expanding these, just make sure that the index that > this implementation depends on is here. > Ok. Will change it to: if (vdev->num_irqs < VFIO_CCW_IO_IRQ_INDEX + 1) { ... ... } > > + > > + argsz = sizeof(*irq_set); > > + irq_info = g_malloc0(argsz); > > + irq_info->index = VFIO_CCW_IO_IRQ_INDEX; > > + irq_info->argsz = argsz; > > + if (ioctl(vdev->fd, VFIO_DEVICE_GET_IRQ_INFO, > > + irq_info) < 0 || irq_info->count < 1) { > > + error_setg(errp, "vfio: Error getting irq info"); > > + goto get_error; > > + } > > + > > + if (event_notifier_init(&vcdev->io_notifier, 0)) { > > + error_setg(errp, "vfio: Unable to init event notifier for IO"); > > + goto get_error; > > + } > > + > > + argsz = sizeof(*irq_set) + sizeof(*pfd); > > + irq_set = g_malloc0(argsz); > > + irq_set->argsz = argsz; > > + irq_set->flags = VFIO_IRQ_SET_DATA_EVENTFD | > > + VFIO_IRQ_SET_ACTION_TRIGGER; > > + irq_set->index = VFIO_CCW_IO_IRQ_INDEX; > > + irq_set->start = 0; > > + irq_set->count = 1; > > + pfd = (int32_t *) &irq_set->data; > > + > > + *pfd = event_notifier_get_fd(&vcdev->io_notifier); > > + qemu_set_fd_handler(*pfd, vfio_ccw_io_notifier_handler, NULL, vcdev); > > + if (ioctl(vdev->fd, VFIO_DEVICE_SET_IRQS, irq_set)) { > > + error_setg(errp, "vfio: Failed to set up io notification"); > > + qemu_set_fd_handler(*pfd, NULL, NULL, vcdev); > > + event_notifier_cleanup(&vcdev->io_notifier); > > + goto set_error; > > + } > > + > > +set_error: > > + g_free(irq_set); > > + > > +get_error: > > + g_free(irq_info); > > +} > > + [...] -- Dong Jia