[comp.unix.i386] ESIX tar problem

trweil@aplcen.apl.jhu.edu (Weil Timothy Robert) (07/10/90)

problem goes like this -
 
       * ftped a socket interface library SI.TAR.Z (50K or so)
       * uncompress fails the 1st time so, check that ulimit = 4096
       * change ulimit = 4000000 & uncompress yields SI.TAR (3.5 Mg)
       * try to examines files with 'tar tvf SI.TAR' & get error
             'directory checksum error'  'bad block' ...
 
whats next?
 
Followup question - what is meaning of bootup dfspace message
 
       real memory = 3800000 (or thereabouts)
    available memory = 2200000 (or thereabouts)
 
    * am I not getting full mileage out of 4 MB of RAM?
    * is dfspace producing this message?
 
Thanks in advance
    trw@aplcen.apl.jhu.edu
    'No such thing as a free lunch, except sometimes'

rick@tmiuv0.uucp (07/12/90)

In article <5923@aplcen.apl.jhu.edu>, trweil@aplcen.apl.jhu.edu (Weil Timothy Robert) writes:
> problem goes like this -
>  
>        * ftped a socket interface library SI.TAR.Z (50K or so)
>        * uncompress fails the 1st time so, check that ulimit = 4096
>        * change ulimit = 4000000 & uncompress yields SI.TAR (3.5 Mg)
>        * try to examines files with 'tar tvf SI.TAR' & get error
>              'directory checksum error'  'bad block' ...
>  
> whats next?

I can't answer that one...try EMailing uunet!zardoz!everex!esixtech or calling
1-714-259-3000 (ESIX Tech Support)

> Followup question - what is meaning of bootup dfspace message
>  
>        real memory = 3800000 (or thereabouts)
>     available memory = 2200000 (or thereabouts)
>  
>     * am I not getting full mileage out of 4 MB of RAM?

Well, the system can't use the area between 640K and 1M, because that's where
lots of BIOS stuff and video boards live.  That explains the "real" memory
difference.  The "available memory" (I think) refers to the amount of free
memory (from the "real" memory) that remains after the Unix kernal is loaded.
I could be wrong.

>     * is dfspace producing this message?

Nope, that's BOOT telling you the memory.  dfspace tells you the amount of
free filespace you have, and that's after all of your filesystems are mounted.
 
> Thanks in advance
>     trw@aplcen.apl.jhu.edu
>     'No such thing as a free lunch, except sometimes'
 
----------------------------------------------------------------------------
[- O] Rick Stevens
  ?   EMail: uunet!zardoz!tmiuv0!rick -or- uunet!zardoz!xyclone!sysop
  V   CIS: 75006,1355 (75006.1355@compuserve.com from Internet)

"I'm tellin' ya, Valiant!  Da whole ting stinks like yesterday's diapers!"
                                - Baby Herman in "Who Framed Roger Rabbit"
----------------------------------------------------------------------------

src@scuzzy.mbx.sub.org (Heiko Blume) (07/22/90)

rick@tmiuv0.uucp writes:
>>     * am I not getting full mileage out of 4 MB of RAM?

>Well, the system can't use the area between 640K and 1M, because that's where
>lots of BIOS stuff and video boards live.  That explains the "real" memory
>difference.  The "available memory" (I think) refers to the amount of free
>memory (from the "real" memory) that remains after the Unix kernal is loaded.
>I could be wrong.

correct. however it might be worth it to have a close look at the things
you can tell your chip set! older chip sets can not map the 384KB RAM
that are physically present (but unused) to addresses at the end of the
physical address space. newer chip sets like the NEAT things *can* do that.
however, you have to consider if you prefer to use shadowing your BIOS ROMs
(might be worthwhile for VGA cards? really don't know if, for example, X11
servers call the BIOS routines or talk to the cards directly) or getting some
more RAM for processes.
-- 
Heiko Blume c/o Diakite   blume@scuzzy.mbx.sub.org    FAX   (+49 30) 882 50 65
Kottbusser Damm 28        blume@netmbx.UUCP           VOICE (+49 30) 691 88 93
D-1000 Berlin 61          blume@netmbx.de             TELEX 184174 intro d
                    "Have you bugged your source today?"