From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
To: Borislav Petkov <bp@alien8.de>
Cc: linux-kernel@vger.kernel.org, H Peter Anvin <hpa@zytor.com>
Subject: Re: [PATCH 6/8] x86, microcode, intel: use cpuid explicitly instead of sync_core
Date: Fri, 7 Nov 2014 16:40:00 -0200 [thread overview]
Message-ID: <20141107184000.GB18128@khazad-dum.debian.net> (raw)
In-Reply-To: <20141107175645.GC5180@pd.tnic>
On Fri, 07 Nov 2014, Borislav Petkov wrote:
> On Mon, Sep 08, 2014 at 02:37:52PM -0300, Henrique de Moraes Holschuh wrote:
> > The protocol to safely read MSR 8BH, described in the Intel SDM vol 3A,
> > section 9.11.7.1, explicitly determines that cpuid with EAX=1 must be
> > used between the wrmsr(0x8B, 0); and the rdmsr(0x8B).
> >
> > The microcode driver was abusing sync_core() to do this, probably
> > because it predates by nearly a decade the current "asm volatile
> > (:::"memory")" implementation of native_cpuid(), which is required for
> > the Intel MSR 8BH access protocol.
>
> Huh, what? Have you taken a look at sync_core() first?
Yes, I did.
> > sync_core() semanthics are that of being a speculative execution
> > barrier, and not "run cpuid with EAX=1".
>
> Again, what?
sync_core() is a speculative execution barrier. That's what it is
documented to do. That's the reason _every_ caller other than the microcode
drivers call it.
In i486, sync_core() does a jmp.
In i586 and above, and x86-64, sync_core() does a cpuid(1).
sync_core() doesn't expect that its callers really want a cpuid(1). If we
ever get a reason to use some other way to insert a speculative execution
barrier, sync_core() is likely to switch to it.
> What is the problem again?
No real problem, other than the fact that the microcode drivers call
sync_core() for what might as well be considered an internal implementation
detail of sync_core().
--
"One disk to rule them all, One disk to find them. One disk to bring
them all and in the darkness grind them. In the Land of Redmond
where the shadows lie." -- The Silicon Valley Tarot
Henrique Holschuh
next prev parent reply other threads:[~2014-11-07 18:40 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-09-08 17:37 [PATCH 0/8] x86, microcode, intel: fixes and enhancements Henrique de Moraes Holschuh
2014-09-08 17:37 ` [PATCH 1/8] x86, microcode, intel: forbid some incorrect metadata Henrique de Moraes Holschuh
2014-10-05 17:34 ` Borislav Petkov
2014-10-05 19:37 ` Henrique de Moraes Holschuh
2014-10-05 21:13 ` Borislav Petkov
2014-10-05 21:49 ` Henrique de Moraes Holschuh
2014-10-06 5:15 ` Borislav Petkov
2014-09-08 17:37 ` [PATCH 2/8] x86, microcode, intel: don't update each HT core twice Henrique de Moraes Holschuh
2014-10-20 13:32 ` Borislav Petkov
2014-10-20 18:24 ` Henrique de Moraes Holschuh
2014-10-28 17:31 ` Borislav Petkov
2014-10-31 18:43 ` Henrique de Moraes Holschuh
2014-11-01 11:06 ` Borislav Petkov
2014-11-01 19:20 ` Henrique de Moraes Holschuh
2014-11-04 15:53 ` Borislav Petkov
2014-09-08 17:37 ` [PATCH 3/8] x86, microcode, intel: clarify log messages Henrique de Moraes Holschuh
2014-10-20 13:52 ` Borislav Petkov
2014-10-21 14:13 ` Henrique de Moraes Holschuh
2014-10-29 9:54 ` Borislav Petkov
2014-10-31 20:08 ` Henrique de Moraes Holschuh
2014-11-07 17:37 ` Borislav Petkov
2014-09-08 17:37 ` [PATCH 4/8] x86, microcode, intel: add error logging to early update driver Henrique de Moraes Holschuh
2014-10-20 15:08 ` Borislav Petkov
2014-10-21 14:10 ` Henrique de Moraes Holschuh
2014-10-30 17:41 ` Borislav Petkov
2014-10-30 18:15 ` Joe Perches
2014-10-31 20:10 ` Henrique de Moraes Holschuh
2014-09-08 17:37 ` [PATCH 5/8] x86, microcode, intel: don't check extsig entry checksum Henrique de Moraes Holschuh
2014-10-30 20:25 ` Borislav Petkov
2014-10-31 17:14 ` Henrique de Moraes Holschuh
2014-11-07 17:49 ` Borislav Petkov
2014-09-08 17:37 ` [PATCH 6/8] x86, microcode, intel: use cpuid explicitly instead of sync_core Henrique de Moraes Holschuh
2014-11-07 17:56 ` Borislav Petkov
2014-11-07 18:40 ` Henrique de Moraes Holschuh [this message]
2014-11-07 19:48 ` Borislav Petkov
2014-09-08 17:37 ` [PATCH 7/8] x86, microcode, intel: guard against misaligned microcode data Henrique de Moraes Holschuh
2014-09-18 0:48 ` Henrique de Moraes Holschuh
2014-11-07 19:59 ` Borislav Petkov
2014-11-07 22:54 ` Henrique de Moraes Holschuh
2014-11-07 23:48 ` Borislav Petkov
2014-11-08 21:57 ` Henrique de Moraes Holschuh
2014-11-11 10:47 ` Borislav Petkov
2014-11-11 16:57 ` Henrique de Moraes Holschuh
2014-11-11 17:13 ` Borislav Petkov
2014-11-11 19:54 ` Henrique de Moraes Holschuh
2014-11-12 12:31 ` Borislav Petkov
2014-11-13 0:18 ` Henrique de Moraes Holschuh
2014-11-13 11:53 ` Borislav Petkov
2014-11-15 23:10 ` Henrique de Moraes Holschuh
2014-11-24 17:35 ` Borislav Petkov
2014-11-25 13:29 ` Henrique de Moraes Holschuh
2014-09-08 17:37 ` [PATCH 8/8] x86, microcode, intel: defend apply_microcode_intel with BUG_ON Henrique de Moraes Holschuh
2014-11-07 20:05 ` Borislav Petkov
2014-11-07 22:56 ` Henrique de Moraes Holschuh
2014-11-07 23:48 ` Borislav Petkov
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=20141107184000.GB18128@khazad-dum.debian.net \
--to=hmh@hmh.eng.br \
--cc=bp@alien8.de \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
/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;
as well as URLs for NNTP newsgroup(s).