X-Authentication-Warning:	delorie.com: mailnull set sender to djgpp-workers-bounces using -f
Date:	Mon, 10 Dec 2001 13:13:16 +0200 (IST)
From:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender:	eliz AT is
To:	Andrew Cottrell <acottrel AT ihug DOT com DOT au>
cc:	Charles Sandmann <sandmann AT clio DOT rice DOT edu>, djgpp-workers AT delorie DOT com
Subject:	Re: v2.03 refresh ready for review/testing
In-Reply-To:	<005201c18165$03a6fd40$0102a8c0@acceleron>
Message-ID:	<Pine.SUN.3.91.1011210130558.21947A-100000@is>
MIME-Version:	1.0
Reply-To:	djgpp-workers AT delorie DOT com
Errors-To:	nobody AT delorie DOT com
X-Mailing-List:	djgpp-workers AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Mon, 10 Dec 2001, Andrew Cottrell wrote:

>   Name                Size in Decimal       Size in Hex
> -------------      ---------------------   -------------
>   MSDOS              12352      ( 12.1K)       3040
>   KBD                 3296      (  3.2K)        CE0
>   HIMEM               1248      (  1.2K)        4E0
>   COMMAND             4768      (  4.7K)       12A0
>   go32-v2            39024      ( 38.1K)       9870
>   go32-v2            42448      ( 41.5K)       A5D0
>   go32-v2            42464      ( 41.5K)       A5E0
>   go32-v2            42496      ( 41.5K)       A600
>   go32-v2            42496      ( 41.5K)       A600
>   go32-v2            42496      ( 41.5K)       A600
>   go32-v2            42464      ( 41.5K)       A5E0
>   go32-v2            42496      ( 41.5K)       A600
>   go32-v2            38992      ( 38.1K)       9850
>   go32-v2            38992      ( 38.1K)       9850
>   go32-v2            39024      ( 38.1K)       9870
>   go32-v2            39024      ( 38.1K)       9870
>   FREE                  32      (  0.0K)         20

This clearly shows that some memory used up by go32-v2 is not released 
when it exits.  (I's also possible that go32-v2 doesn't exit, but that 
sounds unlikely.)

Charles, does ~40KB of low memory ring a bell?  I cannot figure out why 
would go32-v2 need more than twice the size of the transfer buffer in 
conventional memory...

Hmm.. according to this:

>   018C10      go32-v2      004100     Program

it's code, not data.  However, it's possible that "mem/d" thinks it's a 
program because we reuse the stub loader for transfer buffer.

(In case it isn't clear, the reason that go32-v2 is used is probably 
because the GCC build process invokes programs without an explicit .exe 
extension, which causes dosexec to try the file without extension first, 
so it runs the raw COFF executable which requires go32-v2.)

- Raw text -