mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Nicolas Pitre <nicolas.pitre@linaro.org>
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Dave Mielke <Dave@mielke.cc>, linux-kernel@vger.kernel.org
Subject: [PATCH 5/6] vcs: poll(): cope with a deallocated vt
Date: Tue,  8 Jan 2019 22:55:03 -0500	[thread overview]
Message-ID: <20190109035504.8413-6-nicolas.pitre@linaro.org> (raw)
In-Reply-To: <20190109035504.8413-1-nicolas.pitre@linaro.org>

When VT_DISALLOCATE is used on a vt, user space waiting with poll() on
the corresponding /dev/vcs device is not awakened. This is now fixed by
returning POLLHUP|POLLERR to user space.

Also, in the normal screen update case, we don't set POLLERR anymore as
POLLPRI alone is a much more logical response in a non-error situation,
saving some confusion on the user space side. The only known user app
making use of poll() on /dev/vcs* is BRLTTY which is known to cope with
that change already, so the risk of breakage is pretty much nonexistent.

Signed-off-by: Nicolas Pitre <nico@linaro.org>
---
 drivers/tty/vt/vc_screen.c | 21 +++++++++++++++------
 1 file changed, 15 insertions(+), 6 deletions(-)

diff --git a/drivers/tty/vt/vc_screen.c b/drivers/tty/vt/vc_screen.c
index 3dba60825c..1bbe2a30cd 100644
--- a/drivers/tty/vt/vc_screen.c
+++ b/drivers/tty/vt/vc_screen.c
@@ -80,7 +80,7 @@
 struct vcs_poll_data {
 	struct notifier_block notifier;
 	unsigned int cons_num;
-	bool seen_last_update;
+	int event;
 	wait_queue_head_t waitq;
 	struct fasync_struct *fasync;
 };
@@ -94,7 +94,7 @@ vcs_notifier(struct notifier_block *nb, unsigned long code, void *_param)
 		container_of(nb, struct vcs_poll_data, notifier);
 	int currcons = poll->cons_num;
 
-	if (code != VT_UPDATE)
+	if (code != VT_UPDATE && code != VT_DEALLOCATE)
 		return NOTIFY_DONE;
 
 	if (currcons == 0)
@@ -104,7 +104,7 @@ vcs_notifier(struct notifier_block *nb, unsigned long code, void *_param)
 	if (currcons != vc->vc_num)
 		return NOTIFY_DONE;
 
-	poll->seen_last_update = false;
+	poll->event = code;
 	wake_up_interruptible(&poll->waitq);
 	kill_fasync(&poll->fasync, SIGIO, POLL_IN);
 	return NOTIFY_OK;
@@ -261,7 +261,7 @@ vcs_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 
 	poll = file->private_data;
 	if (count && poll)
-		poll->seen_last_update = true;
+		poll->event = 0;
 	read = 0;
 	ret = 0;
 	while (count) {
@@ -616,12 +616,21 @@ static __poll_t
 vcs_poll(struct file *file, poll_table *wait)
 {
 	struct vcs_poll_data *poll = vcs_poll_data_get(file);
-	__poll_t ret = DEFAULT_POLLMASK|EPOLLERR|EPOLLPRI;
+	__poll_t ret = DEFAULT_POLLMASK|EPOLLERR;
 
 	if (poll) {
 		poll_wait(file, &poll->waitq, wait);
-		if (poll->seen_last_update)
+		switch (poll->event) {
+		case VT_UPDATE:
+			ret = DEFAULT_POLLMASK|EPOLLPRI;
+			break;
+		case VT_DEALLOCATE:
+			ret = DEFAULT_POLLMASK|EPOLLHUP|EPOLLERR;
+			break;
+		case 0:
 			ret = DEFAULT_POLLMASK;
+			break;
+		}
 	}
 	return ret;
 }
-- 
2.20.1


  parent reply	other threads:[~2019-01-09  4:04 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-09  3:54 [PATCH 0/6] assorted console/vt/vcs fixes Nicolas Pitre
2019-01-09  3:54 ` [PATCH 1/6] vt: make vt_console_print() compatible with the unicode screen buffer Nicolas Pitre
2019-01-09  3:55 ` [PATCH 2/6] vt: always call notifier with the console lock held Nicolas Pitre
2019-01-09  3:55 ` [PATCH 3/6] vt: invoke notifier on screen size change Nicolas Pitre
2019-01-09  3:55 ` [PATCH 4/6] vcsa: clamp header values when they don't fit Nicolas Pitre
2019-01-09  3:55 ` Nicolas Pitre [this message]
2019-01-10  0:13   ` [PATCH 5/6] vcs: poll(): cope with a deallocated vt Nicolas Pitre
2019-01-09  3:55 ` [PATCH 6/6] vcs: fasync(): make it consistent with poll() Nicolas Pitre
2019-01-18 12:49 ` [PATCH 0/6] assorted console/vt/vcs fixes Greg Kroah-Hartman

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=20190109035504.8413-6-nicolas.pitre@linaro.org \
    --to=nicolas.pitre@linaro.org \
    --cc=Dave@mielke.cc \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.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

all inboxes | Powered by JetHome®