[dpdk-dev] [PATCH v3 1/2] eal: add uevent api for hot plug

[Jeff Guo]

From: "Guo, Jia" <jia.guo at intel.com>

This patch aim to add a variable "uevent_fd" in structure
"rte_intr_handle" for enable kernel object uevent monitoring,
and add some uevent API in rte eal interrupt, that is
“rte_uevent_connect” and “rte_uevent_get”, so that all driver
could use these API to monitor and read out the uevent, then
corresponding to handle these uevent, such as detach or attach
the device.

Signed-off-by: Guo, Jia <jia.guo at intel.com>
---
v3->v2: refine some return error
	refine the string searching logic to aviod memory issue
---
 lib/librte_eal/common/eal_common_pci_uio.c         |   6 +-
 lib/librte_eal/linuxapp/eal/eal_interrupts.c       | 136 ++++++++++++++++++++-
 lib/librte_eal/linuxapp/eal/eal_pci_uio.c          |   6 +
 .../linuxapp/eal/include/exec-env/rte_interrupts.h |  37 ++++++
 4 files changed, 182 insertions(+), 3 deletions(-)

diff --git a/lib/librte_eal/common/eal_common_pci_uio.c b/lib/librte_eal/common/eal_common_pci_uio.c
index 367a681..5b62f70 100644
--- a/lib/librte_eal/common/eal_common_pci_uio.c
+++ b/lib/librte_eal/common/eal_common_pci_uio.c
@@ -117,6 +117,7 @@
 
 	dev->intr_handle.fd = -1;
 	dev->intr_handle.uio_cfg_fd = -1;
+	dev->intr_handle.uevent_fd = -1;
 	dev->intr_handle.type = RTE_INTR_HANDLE_UNKNOWN;
 
 	/* secondary processes - use already recorded details */
@@ -227,7 +228,10 @@
 		close(dev->intr_handle.uio_cfg_fd);
 		dev->intr_handle.uio_cfg_fd = -1;
 	}
-
+	if (dev->intr_handle.uevent_fd >= 0) {
+		close(dev->intr_handle.uevent_fd);
+		dev->intr_handle.uevent_fd = -1;
+	}
 	dev->intr_handle.fd = -1;
 	dev->intr_handle.type = RTE_INTR_HANDLE_UNKNOWN;
 }
diff --git a/lib/librte_eal/linuxapp/eal/eal_interrupts.c b/lib/librte_eal/linuxapp/eal/eal_interrupts.c
index 2e3bd12..2c4a3fb 100644
--- a/lib/librte_eal/linuxapp/eal/eal_interrupts.c
+++ b/lib/librte_eal/linuxapp/eal/eal_interrupts.c
@@ -65,6 +65,10 @@
 #include <rte_errno.h>
 #include <rte_spinlock.h>
 
+#include <sys/socket.h>
+#include <linux/netlink.h>
+#include <sys/epoll.h>
+
 #include "eal_private.h"
 #include "eal_vfio.h"
 #include "eal_thread.h"
@@ -669,10 +673,13 @@ struct rte_intr_source {
 			RTE_SET_USED(r);
 			return -1;
 		}
+
 		rte_spinlock_lock(&intr_lock);
 		TAILQ_FOREACH(src, &intr_sources, next)
-			if (src->intr_handle.fd ==
-					events[n].data.fd)
+			if ((src->intr_handle.fd ==
+					events[n].data.fd) ||
+				(src->intr_handle.uevent_fd ==
+					events[n].data.fd))
 				break;
 		if (src == NULL){
 			rte_spinlock_unlock(&intr_lock);
@@ -858,7 +865,24 @@ static __attribute__((noreturn)) void *
 			}
 			else
 				numfds++;
+
+			/**
+			 * add device uevent file descriptor
+			 * into wait list for uevent monitoring.
+			 */
+			ev.events = EPOLLIN | EPOLLPRI | EPOLLRDHUP | EPOLLHUP;
+			ev.data.fd = src->intr_handle.uevent_fd;
+			if (epoll_ctl(pfd, EPOLL_CTL_ADD,
+					src->intr_handle.uevent_fd, &ev) < 0){
+				rte_panic("Error adding uevent_fd %d epoll_ctl"
+					", %s\n",
+					src->intr_handle.uevent_fd,
+					strerror(errno));
+			} else
+				numfds++;
 		}
+
+
 		rte_spinlock_unlock(&intr_lock);
 		/* serve the interrupt */
 		eal_intr_handle_interrupts(pfd, numfds);
@@ -1255,3 +1279,111 @@ static __attribute__((noreturn)) void *
 
 	return 0;
 }
+
+int
+rte_uevent_connect(void)
+{
+	struct sockaddr_nl addr;
+	int ret;
+	int netlink_fd = -1;
+	int size = 64 * 1024;
+	int nonblock = 1;
+	memset(&addr, 0, sizeof(addr));
+	addr.nl_family = AF_NETLINK;
+	addr.nl_pid = 0;
+	addr.nl_groups = 0xffffffff;
+
+	netlink_fd = socket(PF_NETLINK, SOCK_DGRAM, NETLINK_KOBJECT_UEVENT);
+	if (netlink_fd < 0)
+		return -1;
+
+	setsockopt(netlink_fd, SOL_SOCKET, SO_RCVBUFFORCE, &size, sizeof(size));
+
+	ret = ioctl(netlink_fd, FIONBIO, &nonblock);
+	if (ret != 0) {
+		RTE_LOG(ERR, EAL,
+		"ioctl(FIONBIO) failed\n");
+		close(netlink_fd);
+		return -1;
+	}
+
+	if (bind(netlink_fd, (struct sockaddr *) &addr, sizeof(addr)) < 0) {
+		close(netlink_fd);
+		return -1;
+	}
+
+	return netlink_fd;
+}
+
+static int
+parse_event(const char *buf, struct rte_uevent *event)
+{
+	char action[RTE_UEVENT_MSG_LEN];
+	char subsystem[RTE_UEVENT_MSG_LEN];
+	char dev_path[RTE_UEVENT_MSG_LEN];
+	int i = 0;
+
+	memset(action, 0, RTE_UEVENT_MSG_LEN);
+	memset(subsystem, 0, RTE_UEVENT_MSG_LEN);
+	memset(dev_path, 0, RTE_UEVENT_MSG_LEN);
+
+	while (i < RTE_UEVENT_MSG_LEN) {
+		for (; i < RTE_UEVENT_MSG_LEN; i++) {
+			if (*buf)
+				break;
+			buf++;
+		}
+		if (!strncmp(buf, "ACTION=", 7)) {
+			buf += 7;
+			i += 7;
+			snprintf(action, sizeof(action), "%s", buf);
+		} else if (!strncmp(buf, "DEVPATH=", 8)) {
+			buf += 8;
+			i += 8;
+			snprintf(dev_path, sizeof(dev_path), "%s", buf);
+		} else if (!strncmp(buf, "SUBSYSTEM=", 10)) {
+			buf += 10;
+			i += 10;
+			snprintf(subsystem, sizeof(subsystem), "%s", buf);
+		}
+		for (; i < RTE_UEVENT_MSG_LEN; i++) {
+			if (*buf == '\0')
+				break;
+			buf++;
+		}
+	}
+
+	if (!strncmp(subsystem, "uio", 3)) {
+
+		event->subsystem = RTE_UEVENT_SUBSYSTEM_UIO;
+		if (!strncmp(action, "add", 3))
+			event->action = RTE_UEVENT_ADD;
+		if (!strncmp(action, "remove", 6))
+			event->action = RTE_UEVENT_REMOVE;
+		return 0;
+	}
+
+	return -1;
+}
+
+int
+rte_uevent_get(int fd, struct rte_uevent *uevent)
+{
+	int ret;
+	char buf[RTE_UEVENT_MSG_LEN];
+
+	memset(uevent, 0, sizeof(struct rte_uevent));
+	memset(buf, 0, RTE_UEVENT_MSG_LEN);
+
+	ret = recv(fd, buf, RTE_UEVENT_MSG_LEN - 1, MSG_DONTWAIT);
+	if (ret > 0)
+		return parse_event(buf, uevent);
+	else if (ret < 0) {
+		RTE_LOG(ERR, EAL,
+		"Socket read error(%d): %s\n",
+		errno, strerror(errno));
+		return -1;
+	} else
+		/* connection closed */
+		return -1;
+}
diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
index fa10329..eae9cd5 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
+++ b/lib/librte_eal/linuxapp/eal/eal_pci_uio.c
@@ -231,6 +231,10 @@
 		close(dev->intr_handle.uio_cfg_fd);
 		dev->intr_handle.uio_cfg_fd = -1;
 	}
+	if (dev->intr_handle.uevent_fd >= 0) {
+		close(dev->intr_handle.uevent_fd);
+		dev->intr_handle.uevent_fd = -1;
+	}
 	if (dev->intr_handle.fd >= 0) {
 		close(dev->intr_handle.fd);
 		dev->intr_handle.fd = -1;
@@ -276,6 +280,8 @@
 		goto error;
 	}
 
+	dev->intr_handle.uevent_fd = rte_uevent_connect();
+
 	if (dev->kdrv == RTE_KDRV_IGB_UIO)
 		dev->intr_handle.type = RTE_INTR_HANDLE_UIO;
 	else {
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 6daffeb..0b31a22 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
@@ -90,6 +90,7 @@ struct rte_intr_handle {
 					for uio_pci_generic */
 	};
 	int fd;	 /**< interrupt event file descriptor */
+	int uevent_fd;	 /**< uevent file descriptor */
 	enum rte_intr_handle_type type;  /**< handle type */
 	uint32_t max_intr;             /**< max interrupt requested */
 	uint32_t nb_efd;               /**< number of available efd(event fd) */
@@ -99,6 +100,19 @@ struct rte_intr_handle {
 	int *intr_vec;                 /**< intr vector number array */
 };
 
+#define RTE_UEVENT_MSG_LEN 4096
+#define RTE_UEVENT_SUBSYSTEM_UIO 1
+
+enum rte_uevent_action {
+	RTE_UEVENT_ADD = 0,		/**< uevent type of device add */
+	RTE_UEVENT_REMOVE = 1,	/**< uevent type of device remove*/
+};
+
+struct rte_uevent {
+	enum rte_uevent_action action;	/**< uevent action type */
+	int subsystem;				/**< subsystem id */
+};
+
 #define RTE_EPOLL_PER_THREAD        -1  /**< to hint using per thread epfd */
 
 /**
@@ -236,4 +250,27 @@ struct rte_intr_handle {
 int
 rte_intr_cap_multiple(struct rte_intr_handle *intr_handle);
 
+/**
+ * It read out the uevent from the specific file descriptor.
+ *
+ * @param fd
+ *   The fd which the uevent associated to
+ * @param uevent
+ *   Pointer to the uevent which read from the monitoring fd.
+ * @return
+ *   - On success, zero.
+ *   - On failure, a negative value.
+ */
+int
+rte_uevent_get(int fd, struct rte_uevent *uevent);
+
+/**
+ * Connect to the device uevent file descriptor.
+ * @return
+ *   - On success, the connected uevent fd.
+ *   - On failure, a negative value.
+ */
+int
+rte_uevent_connect(void);
+
 #endif /* _RTE_LINUXAPP_INTERRUPTS_H_ */
-- 
1.8.3.1