linux-kernel - Re: [PATCH] x86 ACPI: normalize segment descriptor register on resume

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <200806261035.19637.rjw@sisk.pl>
Date:	Thu, 26 Jun 2008 10:35:18 +0200
From:	"Rafael J. Wysocki" <rjw@...k.pl>
To:	Andrew Morton <akpm@...ux-foundation.org>
Cc:	Ingo Molnar <mingo@...e.hu>, Pavel Machek <pavel@...e.cz>,
	Len Brown <lenb@...nel.org>,
	ACPI Devel Maling List <linux-acpi@...r.kernel.org>,
	LKML <linux-kernel@...r.kernel.org>,
	pm list <linux-pm@...ts.linux-foundation.org>
Subject: Re: [PATCH] x86 ACPI: normalize segment descriptor register on resume

On Thursday, 26 of June 2008, Andrew Morton wrote:
> On Wed, 25 Jun 2008 22:17:26 +0200 "Rafael J. Wysocki" <rjw@...k.pl> wrote:
> 
> > On Wednesday, 25 of June 2008, Ingo Molnar wrote:
> > > 
> > > * Pavel Machek <pavel@...e.cz> wrote:
> > > 
> > > > Acked-by: Pavel Machek <pavel@...e.cz>
> > > > 
> > > > If this could be sneaked into Ingo's tree for some automated testing, 
> > > > that would be good.
> > > 
> > > sure - i have applied it to tip/out-of-tree. I'm equally nervous about 
> > > this change - it affects every suspend+resume cycle that people do on 
> > > those boxes which are working just fine currently.
> > > 
> > > btw., it would get a lot more coverage on my test-systems if this commit 
> > > in tip/out-of-tree:
> > > 
> > > | commit 01259383c345d13b70efcc549439927ae64dc66d
> > > | Author: David Brownell <david-b@...bell.net>
> > > | Date:   Fri May 16 10:12:36 2008 +0200
> > > |
> > > |     sleepy linux self-test
> > > 
> > > was upstream and if it was enabled more prominently, instead of hidden 
> > > behind the rather obscure condition of:
> > > 
> > >   config PM_TEST_SUSPEND
> > >         bool "Test suspend/resume and wakealarm during bootup"
> > >         depends on SUSPEND && PM_DEBUG && RTC_LIB=y
> > > 
> > > and even then it needs certain other config options related to RTC_LIB 
> > > to actually work during bootup.
> > > 
> > > As a result of all this obstruction, the automated testing i do, which 
> > > builds and boots more than 1 random kernel per minute, will only run 
> > > this self-test once every hour or so.
> > > 
> > > I dont mind if this option breaks boxes (that its purpose: it does the 
> > > same thing that a real suspend+resume does and suspend+resume frequently 
> > > breaks boxes), but right now it's all obscured so heavily which makes 
> > > automated testing a lot harder than it should be.
> > > 
> > > it would be wonderful if this excellent suspend+resume self-test was 
> > > upstream and was more prominent! :-)
> > 
> > I thought it was in linux-next, wasn't it?
> > 
> 
> Len seems to think that Rafael seems to think that Ingo seems to think
> that this patch broke one of his boxes.
> 
> Is it so?

Yes, it is.

The box is apparently broken, but we've been unable to reproduce the breakage
on other very similar boxes, so far.  Investigation continues.

Thanks,
Rafael
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/