This is the mail archive of the cygwin@sourceware.cygnus.com mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: long pause before programs start


On Wed, Sep 15, 1999 at 02:00:09PM -0400, John Wiersba wrote:
>I have just been looking into installing one of the later versions of
>cygwin1.dll (to overcome the bug with the doubling up of .. and . file name
>components which was screwing up the install of the grep 2.3 package).
>However, all the recent snapshots that I've tried, including as far back as
>July, have a serious problem with a long (2+ seconds) pause before many
>programs run.  This doesn't happen at all with the .dll dated 1/16/99.  Bash
>internal commands (like cd) don't have the pause problem and with some
>programs the pause is shorter. 
>
>In fact, things are so screwed up under the 9/14 snapshot that vim needs to
>get several key presses before it will finish painting the terminal window
>and each command needs a few keypresses after it before it appears to be
>performed.

That would be a *bug* in the 9/14 snapshot.

>I wrote a short script:
>
>#!/bin/ksh
>ls | cat | cat | cat | cat | cat | cat | cat | cat >/dev/null
>
>When I run it in a directory with 21 plain files, it takes 5 seconds to
>return to the prompt.  With the 1/16 .dll, it takes less than 1 second.

It takes less than a second for me with the latest DLL.

>How can I diagnose what's going on?

gdb or strace.

>BTW, I looked in the 5/23 cyginw-inst tarball (which I had previously
>downloaded) for strace but only found strace.h.  Where is strace.exe to be
>found?

Why would you assume that the 5/23 cygwin-inst would work with the 9/14
cygwin DLL?

>I'm including cygcheck -s -v output.  I notice that the mount info seems to
>be screwed up.  Could this be the problem?  However, when I do a mount
>command, I get:
>
>   Device           Directory           Type        Flags
>   D:               /                   user        binmode
>   A:               /a/a                user        binmode
>   C:               /c                  user        binmode
>
>Why would cygcheck's mount information be different than mount's?

That could be another bug.  It would be worthwhile to use the latest
version of cygcheck.

-chris

--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]