Re: kern/47324: ulpt(4) and ulpn(4) loose data

[Marc Balmer <marc%msys.ch@localhost>]

The following reply was made to PR kern/47324; it has been noted by GNATS.

From: Marc Balmer <marc%msys.ch@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: kern/47324: ulpt(4) and ulpn(4) loose data
Date: Fri, 14 Dec 2012 10:01:50 +0100

 Am 14.12.2012 um 09:30 schrieb marc%msys.ch@localhost:
 
 >> Number:         47324
 >> Category:       kern
 >> Synopsis:       ulpt(4) and ulpn(4) loose data
 >> Confidential:   no
 >> Severity:       critical
 >> Priority:       high
 >> Responsible:    kern-bug-people
 >> State:          open
 >> Class:          sw-bug
 >> Submitter-Id:   net
 >> Arrival-Date:   Fri Dec 14 08:30:00 +0000 2012
 >> Originator:     Marc Balmer
 >> Release:        5.1.0 and 5.1.2
 >> Organization:
 > micro systems
 >> Environment:
 > NetBSD xxx 5.1.2_PATCH NetBSD 5.1.2_PATCH (GENERIC) #18: Thu Dec  6 =
 17:06:34 CET 2012  =
 root@netbsd-5-1:/usr/obj/sys/arch/i386/compile/GENERIC i386
 >=20
 >> Description:
 > When open()in /dev/ulpt0 or /dev/ulpn0 and then immediately write()ing =
 data, the first byte gets lost in most cases.  Tested on different =
 hardware and with different (receipt) printers.
 >> How-To-Repeat:
 > Write a small program that open()s /dev/ulpt0 and send e.g. an escape =
 sequence to open a cash drawer.  The cash drawer will not open in most =
 cases, because the the initial byte gets lost.
 >> Fix:
 >=20
 
 Whe I write a NUL byte right after the open, subsequent writes will =
 succeed.