[dpdk-dev] [PATCH v2 07/16] Add support for VFIO interrupts, add VFIO header

[Anatoly Burakov]

Creating code to handle VFIO interrupts in EAL interrupts, and also
adding a header eal_vfio.h.

This header checks two things:
* checks if CONFIG_RTE_EAL_VFIO was enabled during build time
* checks that kernel version is 3.6+ so that DPDK would still compile
  on older kernels despite VFIO compilation being enabled by default.

This header also defines a VFIO_PRESENT macro, which should be used to
conditionally compile all the VFIO code. This is because having
CONFIG_RTE_EAL_VFIO enabled doesn't guarantee that the VFIO support is
compiled in, because we're still dependent on kernel version.

Signed-off-by: Anatoly Burakov <anatoly.burakov at intel.com>
---
 lib/librte_eal/linuxapp/eal/eal_interrupts.c       |  203 +++++++++++++++++++-
 lib/librte_eal/linuxapp/eal/include/eal_vfio.h     |   49 +++++
 .../linuxapp/eal/include/exec-env/rte_interrupts.h |    3 +
 3 files changed, 250 insertions(+), 5 deletions(-)
 create mode 100644 lib/librte_eal/linuxapp/eal/include/eal_vfio.h

diff --git a/lib/librte_eal/linuxapp/eal/eal_interrupts.c b/lib/librte_eal/linuxapp/eal/eal_interrupts.c
index 58e1ddf..cb95e2a 100644
--- a/lib/librte_eal/linuxapp/eal/eal_interrupts.c
+++ b/lib/librte_eal/linuxapp/eal/eal_interrupts.c
@@ -36,7 +36,6 @@
 #include <stdlib.h>
 #include <pthread.h>
 #include <sys/queue.h>
-#include <malloc.h>
 #include <stdarg.h>
 #include <unistd.h>
 #include <string.h>
@@ -44,6 +43,7 @@
 #include <inttypes.h>
 #include <sys/epoll.h>
 #include <sys/signalfd.h>
+#include <sys/ioctl.h>
 
 #include <rte_common.h>
 #include <rte_interrupts.h>
@@ -66,6 +66,7 @@
 #include <rte_spinlock.h>
 
 #include "eal_private.h"
+#include "eal_vfio.h"
 
 #define EAL_INTR_EPOLL_WAIT_FOREVER (-1)
 
@@ -87,6 +88,7 @@ union intr_pipefds{
  */
 union rte_intr_read_buffer {
 	int uio_intr_count;              /* for uio device */
+	uint64_t vfio_intr_count;        /* for vfio device */
 	uint64_t timerfd_num;            /* for timerfd */
 	char charbuf[16];                /* for others */
 };
@@ -119,6 +121,173 @@ static struct rte_intr_source_list intr_sources;
 /* interrupt handling thread */
 static pthread_t intr_thread;
 
+/* VFIO interrupts */
+#ifdef VFIO_PRESENT
+
+#define IRQ_SET_BUF_LEN  (sizeof(struct vfio_irq_set) + sizeof(int))
+
+/* enable legacy (INTx) interrupts */
+static int
+vfio_enable_intx(struct rte_intr_handle *intr_handle) {
+	struct vfio_irq_set * irq_set;
+	char irq_set_buf[IRQ_SET_BUF_LEN];
+	int len, ret;
+	int * fd_ptr;
+
+	len = sizeof(irq_set_buf);
+
+	/* enable INTx */
+	irq_set = (struct vfio_irq_set *) irq_set_buf;
+	irq_set->argsz = len;
+	irq_set->count = 1;
+	irq_set->flags = VFIO_IRQ_SET_DATA_EVENTFD | VFIO_IRQ_SET_ACTION_TRIGGER;
+	irq_set->index = VFIO_PCI_INTX_IRQ_INDEX;
+	irq_set->start = 0;
+	fd_ptr = (int*) &irq_set->data;
+	*fd_ptr = intr_handle->fd;
+
+	ret = ioctl(intr_handle->vfio_dev_fd, VFIO_DEVICE_SET_IRQS, irq_set);
+
+	if (ret) {
+		RTE_LOG(ERR, EAL, "Error enabling INTx interrupts for fd %d\n",
+						intr_handle->fd);
+		return -1;
+	}
+
+	/* unmask INTx after enabling */
+	memset(irq_set, 0, len);
+	len = sizeof(struct vfio_irq_set);
+	irq_set->argsz = len;
+	irq_set->count = 1;
+	irq_set->flags = VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_UNMASK;
+	irq_set->index = VFIO_PCI_INTX_IRQ_INDEX;
+	irq_set->start = 0;
+
+	ret = ioctl(intr_handle->vfio_dev_fd, VFIO_DEVICE_SET_IRQS, irq_set);
+
+	if (ret) {
+		RTE_LOG(ERR, EAL, "Error unmasking INTx interrupts for fd %d\n",
+						intr_handle->fd);
+		return -1;
+	}
+	return 0;
+}
+
+/* disable legacy (INTx) interrupts */
+static int
+vfio_disable_intx(struct rte_intr_handle *intr_handle) {
+	struct vfio_irq_set * irq_set;
+	char irq_set_buf[IRQ_SET_BUF_LEN];
+	int len, ret;
+
+	len = sizeof(struct vfio_irq_set);
+
+	/* mask interrupts before disabling */
+	irq_set = (struct vfio_irq_set *) irq_set_buf;
+	irq_set->argsz = len;
+	irq_set->count = 1;
+	irq_set->flags = VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_UNMASK;
+	irq_set->index = VFIO_PCI_INTX_IRQ_INDEX;
+	irq_set->start = 0;
+
+	ret = ioctl(intr_handle->vfio_dev_fd, VFIO_DEVICE_SET_IRQS, irq_set);
+
+	if (ret) {
+		RTE_LOG(ERR, EAL, "Error unmasking INTx interrupts for fd %d\n",
+						intr_handle->fd);
+		return -1;
+	}
+
+	/* disable INTx*/
+	memset(irq_set, 0, len);
+	irq_set->argsz = len;
+	irq_set->count = 0;
+	irq_set->flags = VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_TRIGGER;
+	irq_set->index = VFIO_PCI_INTX_IRQ_INDEX;
+	irq_set->start = 0;
+
+	ret = ioctl(intr_handle->vfio_dev_fd, VFIO_DEVICE_SET_IRQS, irq_set);
+
+	if (ret) {
+		RTE_LOG(ERR, EAL,
+			"Error disabling INTx interrupts for fd %d\n", intr_handle->fd);
+		return -1;
+	}
+	return 0;
+}
+
+/* enable MSI-X interrupts */
+static int
+vfio_enable_msix(struct rte_intr_handle *intr_handle) {
+	int len, ret;
+	char irq_set_buf[IRQ_SET_BUF_LEN];
+	struct vfio_irq_set * irq_set;
+	int * fd_ptr;
+
+	len = sizeof(irq_set_buf);
+
+	irq_set = (struct vfio_irq_set *) irq_set_buf;
+	irq_set->argsz = len;
+	irq_set->count = 1;
+	irq_set->flags = VFIO_IRQ_SET_DATA_EVENTFD | VFIO_IRQ_SET_ACTION_TRIGGER;
+	irq_set->index = VFIO_PCI_MSIX_IRQ_INDEX;
+	irq_set->start = 0;
+	fd_ptr = (int*) &irq_set->data;
+	*fd_ptr = intr_handle->fd;
+
+	ret = ioctl(intr_handle->vfio_dev_fd, VFIO_DEVICE_SET_IRQS, irq_set);
+
+	if (ret) {
+		RTE_LOG(ERR, EAL, "Error enabling MSI-X interrupts for fd %d\n",
+						intr_handle->fd);
+		return -1;
+	}
+
+	/* manually trigger interrupt to enable it */
+	memset(irq_set, 0, len);
+	len = sizeof(struct vfio_irq_set);
+	irq_set->argsz = len;
+	irq_set->count = 1;
+	irq_set->flags = VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_TRIGGER;
+	irq_set->index = VFIO_PCI_MSIX_IRQ_INDEX;
+	irq_set->start = 0;
+
+	ret = ioctl(intr_handle->vfio_dev_fd, VFIO_DEVICE_SET_IRQS, irq_set);
+
+	if (ret) {
+		RTE_LOG(ERR, EAL, "Error triggering MSI-X interrupts for fd %d\n",
+						intr_handle->fd);
+		return -1;
+	}
+	return 0;
+}
+
+/* disable MSI-X interrupts */
+static int
+vfio_disable_msix(struct rte_intr_handle *intr_handle) {
+	struct vfio_irq_set * irq_set;
+	char irq_set_buf[IRQ_SET_BUF_LEN];
+	int len, ret;
+
+	len = sizeof(struct vfio_irq_set);
+
+	irq_set = (struct vfio_irq_set *) irq_set_buf;
+	irq_set->argsz = len;
+	irq_set->count = 0;
+	irq_set->flags = VFIO_IRQ_SET_DATA_NONE | VFIO_IRQ_SET_ACTION_TRIGGER;
+	irq_set->index = VFIO_PCI_MSIX_IRQ_INDEX;
+	irq_set->start = 0;
+
+	ret = ioctl(intr_handle->vfio_dev_fd, VFIO_DEVICE_SET_IRQS, irq_set);
+
+	if (ret)
+		RTE_LOG(ERR, EAL,
+			"Error disabling MSI-X interrupts for fd %d\n", intr_handle->fd);
+
+	return ret;
+}
+#endif
+
 int
 rte_intr_callback_register(struct rte_intr_handle *intr_handle,
 			rte_intr_callback_fn cb, void *cb_arg)
@@ -276,6 +445,16 @@ rte_intr_enable(struct rte_intr_handle *intr_handle)
 	/* not used at this moment */
 	case RTE_INTR_HANDLE_ALARM:
 		return -1;
+#ifdef VFIO_PRESENT
+	case RTE_INTR_HANDLE_VFIO_MSIX:
+		if (vfio_enable_msix(intr_handle))
+			return -1;
+		break;
+	case RTE_INTR_HANDLE_VFIO_LEGACY:
+		if (vfio_enable_intx(intr_handle))
+			return -1;
+		break;
+#endif
 	/* unknown handle type */
 	default:
 		RTE_LOG(ERR, EAL,
@@ -300,7 +479,7 @@ rte_intr_disable(struct rte_intr_handle *intr_handle)
 	case RTE_INTR_HANDLE_UIO:
 		if (write(intr_handle->fd, &value, sizeof(value)) < 0){
 			RTE_LOG(ERR, EAL,
-				"Error enabling interrupts for fd %d\n",
+				"Error disabling interrupts for fd %d\n",
 							intr_handle->fd);
 			return -1;
 		}
@@ -308,6 +487,16 @@ rte_intr_disable(struct rte_intr_handle *intr_handle)
 	/* not used at this moment */
 	case RTE_INTR_HANDLE_ALARM:
 		return -1;
+#ifdef VFIO_PRESENT
+	case RTE_INTR_HANDLE_VFIO_MSIX:
+		if (vfio_disable_msix(intr_handle))
+			return -1;
+		break;
+	case RTE_INTR_HANDLE_VFIO_LEGACY:
+		if (vfio_disable_intx(intr_handle))
+			return -1;
+		break;
+#endif
 	/* unknown handle type */
 	default:
 		RTE_LOG(ERR, EAL,
@@ -357,10 +546,14 @@ eal_intr_process_interrupts(struct epoll_event *events, int nfds)
 		/* set the length to be read dor different handle type */
 		switch (src->intr_handle.type) {
 		case RTE_INTR_HANDLE_UIO:
-			bytes_read = 4;
+			bytes_read = sizeof(buf.uio_intr_count);
 			break;
 		case RTE_INTR_HANDLE_ALARM:
-			bytes_read = sizeof(uint64_t);
+			bytes_read = sizeof(buf.timerfd_num);
+			break;
+		case RTE_INTR_HANDLE_VFIO_MSIX:
+		case RTE_INTR_HANDLE_VFIO_LEGACY:
+			bytes_read = sizeof(buf.vfio_intr_count);
 			break;
 		default:
 			bytes_read = 1;
@@ -397,7 +590,7 @@ eal_intr_process_interrupts(struct epoll_event *events, int nfds)
 				active_cb.cb_fn(&src->intr_handle,
 					active_cb.cb_arg);
 
-				/*get the lcok back. */
+				/*get the lock back. */
 				rte_spinlock_lock(&intr_lock);
 			}
 		}
diff --git a/lib/librte_eal/linuxapp/eal/include/eal_vfio.h b/lib/librte_eal/linuxapp/eal/include/eal_vfio.h
new file mode 100644
index 0000000..ca4982b
--- /dev/null
+++ b/lib/librte_eal/linuxapp/eal/include/eal_vfio.h
@@ -0,0 +1,49 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of Intel Corporation nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef EAL_VFIO_H_
+#define EAL_VFIO_H_
+
+/*
+ * determine if VFIO is present on the system
+ */
+#ifdef RTE_EAL_VFIO
+#include <linux/version.h>
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(3,6,0)
+#include <linux/vfio.h>
+
+#define VFIO_PRESENT
+#endif /* kernel version */
+#endif /* RTE_EAL_VFIO */
+
+#endif /* EAL_VFIO_H_ */
diff --git a/lib/librte_eal/linuxapp/eal/include/exec-env/rte_interrupts.h b/lib/librte_eal/linuxapp/eal/include/exec-env/rte_interrupts.h
index 6733948..b160efe 100644
--- a/lib/librte_eal/linuxapp/eal/include/exec-env/rte_interrupts.h
+++ b/lib/librte_eal/linuxapp/eal/include/exec-env/rte_interrupts.h
@@ -41,12 +41,15 @@
 enum rte_intr_handle_type {
 	RTE_INTR_HANDLE_UNKNOWN = 0,
 	RTE_INTR_HANDLE_UIO,      /**< uio device handle */
+	RTE_INTR_HANDLE_VFIO_LEGACY,  /**< vfio device handle (legacy) */
+	RTE_INTR_HANDLE_VFIO_MSIX,    /**< uio device handle (MSIX) */
 	RTE_INTR_HANDLE_ALARM,    /**< alarm handle */
 	RTE_INTR_HANDLE_MAX
 };
 
 /** Handle for interrupts. */
 struct rte_intr_handle {
+	int vfio_dev_fd;                 /**< VFIO device file descriptor */
 	int fd;                          /**< file descriptor */
 	enum rte_intr_handle_type type;  /**< handle type */
 };
-- 
1.7.0.7