Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 29 Jan 1996 14:58:36 -0700 (MST)
From:      Terry Lambert <terry@lambert.org>
To:        bde@zeta.org.au (Bruce Evans)
Cc:        dyson@freefall.freebsd.org, terry@lambert.org, current@freefall.freebsd.org
Subject:   Optimization topics
Message-ID:  <199601292158.OAA04701@phaeton.artisoft.com>
In-Reply-To: <199601262317.KAA27798@godzilla.zeta.org.au> from "Bruce Evans" at Jan 27, 96 10:17:54 am

next in thread | previous in thread | raw e-mail | index | archive | help
With all the pipe optimization stuff and all the work I've been doing
in the Windows95 kernel recently. it occurs to me that there might be
room for some optimizations in FreeBSD that Microsoft totally buggered
up in Windows95.

Specifically, it should be possible to run FreeBSD with the unaligned
access bit set to disallow unaligned access of word/dword/qword objects.

At the very least, the kernel code should be clean of this cruft.

In Windows95, some programmers apparently believe that adding 3 bytes
to the end of a packed structure containing a byte followed by 20
longs means "align to a longword boundry" -- at least that's what
the comments in the DDK header files say.

Clearly, these people never ran their own code in a kernel debugger,
or they simply failed to see the significance of unaligned access.

Many of the problems derive from unaligned stacks being passed down
to otherwise "correct" code.  Fie.

Maybe this is a direction that can be explored?


					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199601292158.OAA04701>