PATCH: Constant initial fdtable_size

PATCH: Constant initial fdtable_size

[Bart Schaefer]

This is intended for the 4-0-patches branch; it should fix the crash reported
in 16082.  Presumably it's sufficiently noncontroversial to commit?

Index: Src/init.c
===================================================================
--- Src/init.c	2001/09/24 15:40:12	1.6
+++ Src/init.c	2001/10/22 14:41:42
@@ -1190,7 +1190,8 @@
 	  break;
     } while (zsh_name);
 
-    fdtable_size = zopenmax();
+    /* Not zopenmax() here: it may return a number too big for zcalloc(). */
+    fdtable_size = 256; /* This grows as necessary, see utils.c:movefd(). */
     fdtable = zcalloc(fdtable_size);
 
     createoptiontable();

-- 
Bart Schaefer                                 Brass Lantern Enterprises
http://www.well.com/user/barts              http://www.brasslantern.com

Zsh: http://www.zsh.org | PHPerl Project: http://phperl.sourceforge.net   

Re: PATCH: Constant initial fdtable_size

[Peter Stephenson]

Bart Schaefer wrote:
> This is intended for the 4-0-patches branch; it should fix the crash reported
> in 16082.  Presumably it's sufficiently noncontroversial to commit?

I don't see any problem.

Is there anything left to be done for the next 4.0 release?

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR Ltd., Science Park, Milton Road,
Cambridge, CB4 0WH, UK                          Tel: +44 (0)1223 392070


**********************************************************************
The information transmitted is intended only for the person or
entity to which it is addressed and may contain confidential 
and/or privileged material. 
Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by 
persons or entities other than the intended recipient is 
prohibited.  
If you received this in error, please contact the sender and 
delete the material from any computer.
**********************************************************************

Anything left (Re: PATCH: Constant initial fdtable_size)

[Bart Schaefer]

On Oct 22,  3:57pm, Peter Stephenson wrote:
}
} Is there anything left to be done for the next 4.0 release?

I had been going to try to apply 15354 to the docs, but I think I won't,
as it touches lots of files and doesn't add any essential information.

Should 16110 go onto the branch?

-- 
Bart Schaefer                                 Brass Lantern Enterprises
http://www.well.com/user/barts              http://www.brasslantern.com

Zsh: http://www.zsh.org | PHPerl Project: http://phperl.sourceforge.net   

Re: Anything left (Re: PATCH: Constant initial fdtable_size)

[Bart Schaefer]

On Oct 22,  3:17pm, Bart Schaefer wrote:
}
} Should 16110 go onto the branch?

I meant 16111, of course.

-- 
Bart Schaefer                                 Brass Lantern Enterprises
http://www.well.com/user/barts              http://www.brasslantern.com

Zsh: http://www.zsh.org | PHPerl Project: http://phperl.sourceforge.net   

Re: Anything left (Re: PATCH: Constant initial fdtable_size)

[Oliver Kiddle]

> On Oct 22,  3:57pm, Peter Stephenson wrote:
> }
> } Is there anything left to be done for the next 4.0 release?

There was nothing I had left.

Did anyone look at 16033 (fix for QNX) as if it is a fix it should go in
4.0. I know nothing about rlimits so am not a good person to judge if
the patch should be applied or not.

Oliver

Re: Anything left (Re: PATCH: Constant initial fdtable_size)

[Peter Stephenson]

Oliver Kiddle wrote:
> Did anyone look at 16033 (fix for QNX) as if it is a fix it should go in
> 4.0. I know nothing about rlimits so am not a good person to judge if
> the patch should be applied or not.

It hasn't even gone in 4.1, yet, so I'll apply it there.  It looks
uncontroversial for other systems.  Is anyone using another system where
(RLIMIT_RSS == RLIMIT_VMEM)?

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR Ltd., Science Park, Milton Road,
Cambridge, CB4 0WH, UK                          Tel: +44 (0)1223 392070


**********************************************************************
The information transmitted is intended only for the person or
entity to which it is addressed and may contain confidential 
and/or privileged material. 
Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by 
persons or entities other than the intended recipient is 
prohibited.  
If you received this in error, please contact the sender and 
delete the material from any computer.
**********************************************************************

RE: Anything left (Re: PATCH: Constant initial fdtable_size)

[Borsenkow Andrej]

> Is there anything left to be done for the next 4.0 release?
> 

I am applying 15181 (-KPIC flag) and 15180 (AIX tweaks).

-andrej

RE: Anything left (Re: PATCH: Constant initial fdtable_size)

[Borsenkow Andrej]

I applied 15390 (_mount for Cygwin) it should be harmless.

Should we apply 15278 (preserve-prefix, mostly for // on Cygwin) as
well? In head I accidentally committed it. May it have evil
side-effects? This probably needs documentation then (and in HEAD as
well).

-andrej

PATCH: Infinite loop in `case' with xtrace

[Bart Schaefer]

(Read all the way to the end, there are comments interspersed with patches.)

I just did

zsh% zstyle \* preserve-prefix \~///
zsh% ls ~///<C-x ?>

and _arguments went into an infinite loop.  At first I thought it was this:

Index: Completion/Base/Utility/_arguments
===================================================================
--- Completion/Base/Utility/_arguments	2001/09/05 15:22:28	1.6
+++ Completion/Base/Utility/_arguments	2001/10/23 14:56:53
@@ -183,6 +183,7 @@
   -w)  optarg=yes; shift;;
   -s)  singopt=(-s); shift;;
   -W)  alwopt=arg; shift;;
+  *)   break;;
   esac
 done

But when that didn't fix it I found it was really this, caused by 16041:

Index: Src/loop.c
===================================================================
--- Src/loop.c	2001/10/17 14:38:28	1.3
+++ Src/loop.c	2001/10/23 15:09:50
@@ -553,9 +553,8 @@
 	    printprompt4();
 	    fprintf(xtrerr, "case %s (%s)\n", word, pat2);
 	    fflush(xtrerr);
-	    state->pc++;
-	} else
-	    state->pc += 2;
+	}
+	state->pc += 2;
 
 	if (*spprog != dummy_patprog1 && *spprog != dummy_patprog2)
 	    pprog = *spprog;


I note that neither 16038 nor 16041 has been applied to the 4-0-patches
branch.  Are they not needed there?  (And if 16041 is needed, be sure to
apply the above as well.)

-- 
Bart Schaefer                                 Brass Lantern Enterprises
http://www.well.com/user/barts              http://www.brasslantern.com

Zsh: http://www.zsh.org | PHPerl Project: http://phperl.sourceforge.net