connie@nih-csl.UUCP (connie eppich) (05/29/87)
I am having an intermittent problem with the new Emacs. When I test a
program I will often suspend the Emacs session to run something
in the c-shell outside of Emacs. I have been having Emacs "get stuck"
after I issue the "foreground" command, "fg", to the shell; my terminal
is completely locked up. My only recourse is to disconnect.
It is not an acceptable solution to stay in an Emacs shell buffer because
the dbxdebugger does not always work well in that context. On the other hand,
I don't consider it acceptable to have to exit Emacs each time.
Has anyone else had this problem? Any suggestions?
Connie Eppich
Computer Systems Lab
National Institutes of Health
Bethesda, MDmkhaw@teknowledge-vaxc.ARPA (Michael Khaw) (05/29/87)
In article <223@nih-csl.UUCP> connie@nih-csl.UUCP (connie eppich) writes: >I am having an intermittent problem with the new Emacs. When I test a >program I will often suspend the Emacs session to run something >in the c-shell outside of Emacs. I have been having Emacs "get stuck" >after I issue the "foreground" command, "fg", to the shell; my terminal >is completely locked up. My only recourse is to disconnect. We had this problem under Sun OS 3.0 when using a shelltool window rlogin'ed to a vax and running GNU Emacs 18.33 on the vax. If you used telnet instead of rlogin, there was no problem. Since we've gone to Sun OS 3.2 and GNU Emacs 18.41, the problem has gone away. Other than that, it's been fine (except, of course, if you don't disable the suspend-emacs function when you run emacstool under suntools). Mike Khaw -- internet: mkhaw@teknowledge-vaxc.arpa usenet: {hplabs|sun|ucbvax|decwrl|sri-unix}!mkhaw%teknowledge-vaxc.arpa USnail: Teknowledge Inc, 1850 Embarcadero Rd, POB 10119, Palo Alto, CA 94303
danny@calgary.UUCP (Danny Levinson) (05/30/87)
In article <223@nih-csl.UUCP>, connie@nih-csl.UUCP (connie eppich) writes: > I am having an intermittent problem with the new Emacs. When I test a > program I will often suspend the Emacs session to run something > in the c-shell outside of Emacs. I have been having Emacs "get stuck" > after I issue the "foreground" command, "fg", to the shell; my terminal > is completely locked up. My only recourse is to disconnect. I have indeed had this problem, but not just with gnu emacs. My guess is that you, like myself, are working on a sun (bsd4.2, v3.2?) and that we are experiencing a bug in the Unix tty driver. Perhaps it arises because doing certain "ioctl"s before the tty driver has had a chance to flush the echo of the <return> following the "fg" command has side effects. At any rate, putting a "sleep(1)" before the ioctl in question (and, for safety's sake, after it too) worked for me in Gosling emacs. I haven't actually fixed it in Gnu as (1) it occurs far less regularly (perhaps gosmacs does less work twixt ioctl and generating output), and (2) the ioctl calls are harder to find in the Gnu code. I am given to understand that a fuller description of the Unix bug and a prescription for its cure appeared in SunSpots (or a related group). Try your local guru for a definitive fix! -- danny levinson university of calgary
howie@cunixc (Howie Kaye) (05/30/87)
I have also seen GNUemacs 18.41 hang a job, but it seems to do it
during a suspend. Running sps from another job shows that the shell
is still in a kernel sleep, but emacs is stopped. It looks like a
race condition in the kernel somewhere. this is in Ultrix 1.2, and
only seems to happen under rsh. emacs is the only program that we've
seen this happen with.
------------------------------------------------------------------------
Howie Kaye howie@CUNIXC.Columbia.EDU
Columbia University HKAUS@cuvma (bitnet)
Systems Integration Group {?}!seismo!columbia!cucca!howielauri@MITRE-BEDFORD.ARPA.UUCP (06/02/87)
We are running emacs 18.41 on a VAX 8600. It appears that EDT emulation does not work properly. When we type M-x edt-emulation-on, we get the following message: Loading edt... And then: Symbol's value as variable is void: e Has anyone else seen this before? Is there a fix? Thanks.