Dave,
I did a
rpm -rebuild egcs
rpm -rebuild glibc
ldconfig
ldconfig
and it went away.
I reinstalled a clean Open Linux 2.4 and just did
ldconfig
ldconfig
without rebuilding and it went also went away, so I don't think rebuilding
had much to do with it. I did spend any time looking further into it.
One issue with this is that the jump into the spinning code will always be a
long jump since these segments end up where ever the linker feels like
emitting them. Inlinging the spining code to less than a 512 byte boundry
would make the long jump into a short jump and speed things up. I've also
noticed that the linker does not always align them on 16 bytes boudries
(which is what Intel likes to see for speed).
All assembler functions in NetWare are always preceeded with an 'align 16'
directive and this does increase execution speed -- something to think
about...
:-)
Jeff
----- Original Message -----
From: "David S. Miller" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>
Sent: Tuesday, September 19, 2000 7:41 PM
Subject: Re: NMI Watchdog detected LOCKUP on CPU1
(stext_lock)(2.4.0-test9-pre2)
> Date: Tue, 19 Sep 2000 19:44:30 -0600
> From: "Jeff V. Merkey" <[EMAIL PROTECTED]>
>
> It does not seem to be saving any memory space doing it this way,
> since I've noticed tons of these little segments all over the
> place.
>
> None of them can be eliminated because each one branches back to a
> unique PC.
>
> This independant of whatever linker bug you were running into.
>
> Later,
> David S. Miller
> [EMAIL PROTECTED]
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [EMAIL PROTECTED]
> Please read the FAQ at http://www.tux.org/lkml/
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
Please read the FAQ at http://www.tux.org/lkml/
