[PATCH 1/1] virtio: console: wait for first console port for early console output

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



From: Christian Borntraeger <borntraeger@xxxxxxxxxx>

On s390 I have seen some random

"Warning: unable to open an initial console"

boot failure. Turns out that tty_open fails, because the
hvc_alloc was not yet done. In former times this could not happen,
since the probe function automatically called hvc_alloc. With newer
versions (multiport) some host<->guest interaction is required
before hvc_alloc is called. This might be too late, especially if
an initramfs is involved. Lets use a completion if we have
multiport and an early console.

[Amit:
  * Use NULL instead of 0 for pointer comparison
  * Rename 'port_added' to 'early_console_added'
  * Re-format, re-word commit message
  * Rebase patch on top of current queue]

Signed-off-by: Christian Borntraeger <borntraeger@xxxxxxxxxx>
Signed-off-by: Amit Shah <amit.shah@xxxxxxxxxx>
---
 drivers/char/virtio_console.c |   20 ++++++++++++++++++++
 1 files changed, 20 insertions(+), 0 deletions(-)

diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
index 9ea3b5e..7f2c6e5 100644
--- a/drivers/char/virtio_console.c
+++ b/drivers/char/virtio_console.c
@@ -19,6 +19,7 @@
  */
 #include <linux/cdev.h>
 #include <linux/debugfs.h>
+#include <linux/completion.h>
 #include <linux/device.h>
 #include <linux/err.h>
 #include <linux/freezer.h>
@@ -74,6 +75,7 @@ struct ports_driver_data {
 static struct ports_driver_data pdrvdata;
 
 DEFINE_SPINLOCK(pdrvdata_lock);
+DECLARE_COMPLETION(early_console_added);
 
 /* This struct holds information that's relevant only for console ports */
 struct console {
@@ -1366,6 +1368,7 @@ static void handle_control_message(struct ports_device *portdev,
 			break;
 
 		init_port_console(port);
+		complete(&early_console_added);
 		/*
 		 * Could remove the port here in case init fails - but
 		 * have to notify the host first.
@@ -1668,6 +1671,10 @@ static int __devinit virtcons_probe(struct virtio_device *vdev)
 	struct ports_device *portdev;
 	int err;
 	bool multiport;
+	bool early = early_put_chars != NULL;
+
+	/* Ensure to read early_put_chars now */
+	barrier();
 
 	portdev = kmalloc(sizeof(*portdev), GFP_KERNEL);
 	if (!portdev) {
@@ -1739,6 +1746,19 @@ static int __devinit virtcons_probe(struct virtio_device *vdev)
 
 	__send_control_msg(portdev, VIRTIO_CONSOLE_BAD_ID,
 			   VIRTIO_CONSOLE_DEVICE_READY, 1);
+
+	/*
+	 * If there was an early virtio console, assume that there are no
+	 * other consoles. We need to wait until the hvc_alloc matches the
+	 * hvc_instantiate, otherwise tty_open will complain, resulting in
+	 * a "Warning: unable to open an initial console" boot failure.
+	 * Without multiport this is done in add_port above. With multiport
+	 * this might take some host<->guest communication - thus we have to
+	 * wait.
+	 */
+	if (multiport && early)
+		wait_for_completion(&early_console_added);
+
 	return 0;
 
 free_vqs:
-- 
1.7.6.2

--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]
  Powered by Linux