Bug 163528

Summary:	/dev/tty won't open during blocking /dev/ttyS1 open
Product:	Red Hat Enterprise Linux 4	Reporter:	karl malbrain <malbrain>
Component:	kernel	Assignee:	Jason Baron <jbaron>
Status:	CLOSED ERRATA	QA Contact:	Brian Brock <bbrock>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	4.0	CC:	knoel, poelstra
Target Milestone:	---
Target Release:	---
Hardware:	i386
OS:	Linux
Whiteboard:
Fixed In Version:	RHSA-2005-514	Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2005-10-05 13:43:12 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:
Bug Blocks:	156322

Description karl malbrain 2005-07-18 16:47:34 UTC

From Bugzilla Helper:
User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0)

Description of problem:
One of the Red Hat patches to 2.6.9 moves the tty_sem to encompass uart_open in tty_io.c.  After issuing a "blocking" uart_open (wait for carrier detect on the modem) ALL subsequent opens to any /dev/tty device are blocking on the tty_sem semaphore which is being held during the sleep cycle in uart_open that's waiting for carrier-detect from the modem.

This behaviour was introduced in the Red Hat "bits" patch.  The tty_sem semaphore cannot be held during the call to uart_open.

karl m

Version-Release number of selected component (if applicable):
kernel 2.6.9-11

How reproducible:
Always

Steps to Reproduce:
1. Issue a blocking open to /dev/ttyS1 with CLOCAL cleared and w/o O_NDELAY.
2. Start a shell prompt (this will open /dev/tty)
3. The open of /dev/tty will hang on tty_sem.
  

Actual Results:  Unable to start new sessions or open other major-device 5 inodes.

Expected Results:  The subsequent opens should succeed immediately, rather than block.

Additional info:

Whoever "cleaned up" tty_io.c in the kernel in "2.6.10-ac-selected-bits-patch" made an assumption that the underlying device driver open would not sleep with the semaphore held open.

Comment 1 Jason Baron 2005-07-22 22:05:14 UTC

This issue has been addressed in our laest builds of U2. Please see:
http://people.redhat.com/~jbaron/rhel4/ for a test kernel.

Comment 7 Red Hat Bugzilla 2005-10-05 13:43:12 UTC

An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHSA-2005-514.html