virtualization.lists.linux-foundation.org archive mirror
 help / color / mirror / Atom feed
From: Amit Shah <amit.shah@redhat.com>
To: Rusty Russell <rusty@rustcorp.com.au>
Cc: Virtualization List <virtualization@lists.linux-foundation.org>,
	"Michael S. Tsirkin" <mst@redhat.com>
Subject: Re: [PATCH 10/10] virtio: console: Add support for nonblocking write()s
Date: Tue, 6 Apr 2010 20:24:31 +0530	[thread overview]
Message-ID: <20100406145431.GA2677@amit-x200.redhat.com> (raw)
In-Reply-To: <201004061242.58840.rusty@rustcorp.com.au>

On (Tue) Apr 06 2010 [12:42:58], Rusty Russell wrote:
> On Mon, 5 Apr 2010 11:24:14 pm Amit Shah wrote:
> > If the host port is not open, a write() should either just return if the
> > file is opened in non-blocking mode, or block till the host port is
> > opened.
> > 
> > Also, don't spin till host consumes data for nonblocking ports. For
> > non-blocking ports, we can do away with the spinning and reclaim the
> > buffers consumed by the host on the next write call or on the condition
> > that'll make poll return.
> 
> I'm only reading the patch so I might have missed it, but what's the
> locking going on here?
> 
> Can we race thinking we're full or not full incorrectly?

Turns out just this much on top of this patch should be sufficient.


diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
index 0c632ee..6541f76 100644
--- a/drivers/char/virtio_console.c
+++ b/drivers/char/virtio_console.c
@@ -158,6 +158,9 @@ struct port {
 	 */
 	spinlock_t inbuf_lock;
 
+	/* Protect the operations on the out_vq. */
+	spinlock_t outbuf_lock;
+
 	/* The IO vqs for this port */
 	struct virtqueue *in_vq, *out_vq;
 
@@ -409,12 +412,15 @@ static ssize_t send_control_msg(struct port *port, unsigned int event,
 static void reclaim_consumed_buffers(struct port *port)
 {
 	void *buf;
+	unsigned long flags;
 	unsigned int len;
 
+	spin_lock_irqsave(&port->outbuf_lock, flags);
 	while ((buf = port->out_vq->vq_ops->get_buf(port->out_vq, &len))) {
 		kfree(buf);
 		port->outvq_full = false;
 	}
+	spin_unlock_irqrestore(&port->outbuf_lock, flags);
 }
 
 static ssize_t send_buf(struct port *port, void *in_buf, size_t in_count,
@@ -423,6 +429,7 @@ static ssize_t send_buf(struct port *port, void *in_buf, size_t in_count,
 	struct scatterlist sg[1];
 	struct virtqueue *out_vq;
 	ssize_t ret;
+	unsigned long flags;
 	unsigned int len;
 
 	out_vq = port->out_vq;
@@ -430,6 +437,9 @@ static ssize_t send_buf(struct port *port, void *in_buf, size_t in_count,
 	reclaim_consumed_buffers(port);
 
 	sg_init_one(sg, in_buf, in_count);
+
+	spin_lock_irqsave(&port->outbuf_lock, flags);
+
 	ret = out_vq->vq_ops->add_buf(out_vq, sg, 1, 0, in_buf);
 
 	/* Tell Host to go! */
@@ -437,14 +447,14 @@ static ssize_t send_buf(struct port *port, void *in_buf, size_t in_count,
 
 	if (ret < 0) {
 		in_count = 0;
-		goto fail;
+		goto done;
 	}
 
 	if (ret == 0)
 		port->outvq_full = true;
 
 	if (nonblock)
-		return in_count;
+		goto done;
 
 	/*
 	 * Wait till the host acknowledges it pushed out the data we
@@ -454,7 +464,8 @@ static ssize_t send_buf(struct port *port, void *in_buf, size_t in_count,
 	 */
 	while (!out_vq->vq_ops->get_buf(out_vq, &len))
 		cpu_relax();
-fail:
+done:
+	spin_unlock_irqrestore(&port->outbuf_lock, flags);
 	/*
 	 * We're expected to return the amount of data we wrote -- all
 	 * of it
@@ -1009,6 +1020,7 @@ static int add_port(struct ports_device *portdev, u32 id)
 	}
 
 	spin_lock_init(&port->inbuf_lock);
+	spin_lock_init(&port->outbuf_lock);
 	init_waitqueue_head(&port->waitqueue);
 
 	/* Fill the in_vq with buffers so the host can send us data. */

  parent reply	other threads:[~2010-04-06 14:54 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-04-05 13:54 [PATCH 00/10] (v5) virtio: console: Fixes, new way of discovering ports Amit Shah
2010-04-05 13:54 ` [PATCH 01/10] virtio: console: Add a __send_control_msg() that can send messages without a valid port Amit Shah
2010-04-05 13:54   ` [PATCH 02/10] virtio: console: Let host know of port or device add failures Amit Shah
2010-04-05 13:54     ` [PATCH 03/10] virtio: console: Return -EPIPE to hvc_console if we lost the connection Amit Shah
2010-04-05 13:54       ` [PATCH 04/10] virtio: console: Don't call hvc_remove() on unplugging console ports Amit Shah
2010-04-05 13:54         ` [PATCH 05/10] virtio: console: Remove config work handler Amit Shah
2010-04-05 13:54           ` [PATCH 06/10] virtio: console: Move code around for future patches Amit Shah
2010-04-05 13:54             ` [PATCH 07/10] virtio: console: Use a control message to add ports Amit Shah
2010-04-05 13:54               ` [PATCH 08/10] virtio: console: Don't always create a port 0 if using multiport Amit Shah
2010-04-05 13:54                 ` [PATCH 09/10] virtio: console: Rename wait_is_over() to will_read_block() Amit Shah
2010-04-05 13:54                   ` [PATCH 10/10] virtio: console: Add support for nonblocking write()s Amit Shah
2010-04-06  3:12                     ` Rusty Russell
2010-04-06  3:54                       ` Amit Shah
2010-04-06 14:54                       ` Amit Shah [this message]
2010-04-06  3:09               ` [PATCH 07/10] virtio: console: Use a control message to add ports Rusty Russell

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20100406145431.GA2677@amit-x200.redhat.com \
    --to=amit.shah@redhat.com \
    --cc=mst@redhat.com \
    --cc=rusty@rustcorp.com.au \
    --cc=virtualization@lists.linux-foundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).