From: Linus Torvalds <torvalds@osdl.org>
To: Stas Sergeev <stsp@aknet.ru>
Cc: Pavel Machek <pavel@ucw.cz>, Alan Cox <alan@redhat.com>,
Linux kernel <linux-kernel@vger.kernel.org>,
Petr Vandrovec <vandrove@vc.cvut.cz>,
Denis Vlasenko <vda@port.imtp.ilyichevsk.odessa.ua>
Subject: Re: [patch] x86: fix ESP corruption CPU bug
Date: Sun, 13 Mar 2005 13:13:16 -0800 (PST) [thread overview]
Message-ID: <Pine.LNX.4.58.0503131306450.2822@ppc970.osdl.org> (raw)
In-Reply-To: <4234A8DD.9080305@aknet.ru>
On Sun, 13 Mar 2005, Stas Sergeev wrote:
>
> Such an optimization will cost three more
> instructions, one of which is a "taken"
> jump.
I think Pavel missed the fact that you have to check the VM86 bit in
eflags before you check SS, since otherwise SS doesn't mean anything
special at all (ie checking for just the normal SS isn't correct: you
could have a 16-bit SS that looks normal, but is actually a vm86 segment).
Pavel: for the same reason you have to check the low two bits of CS too,
since if they are zero, then SS hasn't been saved on the stack at all, so
comparing it against some normal value is meaningless.
That said, the "ldt_ss" case should be moved _after_ the conditional
tests, since most CPU's out there will do static prediction based on
forward/backwards direction if the branch predictor isn't primed. And so
since it's an unusual case, the branch should be a forward branch, which
is usually the not-taken one (this depends on the core, of course, and you
could also add the prefix byte to mark it explicitly predict-not-taken for
the newer cores that support it).
Linus
next prev parent reply other threads:[~2005-03-13 21:11 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-03-13 18:20 Stas Sergeev
2005-03-13 18:52 ` Grzegorz Kulewski
2005-03-13 19:11 ` Stas Sergeev
2005-03-13 19:37 ` Ondrej Zary
2005-03-13 19:46 ` Stas Sergeev
2005-03-13 20:02 ` Pavel Machek
2005-03-13 20:10 ` Pavel Machek
2005-03-13 20:55 ` Stas Sergeev
2005-03-13 21:13 ` Linus Torvalds [this message]
2005-03-13 22:06 ` [patch] x86: fix ESP corruption CPU bug (take 2) Stas Sergeev
2005-03-14 19:29 ` Alan Cox
2005-03-14 19:59 ` Stas Sergeev
2005-03-15 3:34 ` Andrew Morton
2005-03-15 10:48 ` x86: spin_unlock(), spin_unlock_irq() & others are out of line ? Eric Dumazet
2005-03-15 19:44 ` Lee Revell
2005-03-15 19:48 ` [patch] x86: fix ESP corruption CPU bug (take 2) Stas Sergeev
2005-03-13 23:17 ` [patch] x86: fix ESP corruption CPU bug Pavel Machek
2005-03-13 23:54 ` Linus Torvalds
2005-03-14 0:16 ` Linus Torvalds
2005-03-14 4:52 ` Stas Sergeev
2005-03-14 9:34 ` Andi Kleen
2005-03-14 10:37 ` [BUG?] x86_64 : Can not read /dev/kmem ? Eric Dumazet
2005-03-21 19:37 ` Andi Kleen
2005-03-22 15:24 ` Andi Kleen
2005-03-14 15:21 ` [patch] x86: fix ESP corruption CPU bug Jakob Eriksson
2005-03-14 17:03 ` linux-os
2005-03-14 17:10 ` Pavel Machek
2005-03-14 19:24 ` Brian Gerst
2005-03-14 20:21 ` Stas Sergeev
2005-03-14 18:02 ` Stas Sergeev
2005-03-14 17:29 ` Stas Sergeev
-- strict thread matches above, loose matches on Subject: below --
2005-03-14 11:10 Zoltan Boszormenyi
2005-01-03 23:39 Stas Sergeev
2005-01-04 0:01 ` Linus Torvalds
2005-01-04 1:58 ` Stas Sergeev
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=Pine.LNX.4.58.0503131306450.2822@ppc970.osdl.org \
--to=torvalds@osdl.org \
--cc=alan@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=pavel@ucw.cz \
--cc=stsp@aknet.ru \
--cc=vandrove@vc.cvut.cz \
--cc=vda@port.imtp.ilyichevsk.odessa.ua \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®