linux-kernel - Re: [tip:perf/core] perf/x86: Clean up cap_user

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: <5252EDBE.7030805@zytor.com>
Date:	Mon, 07 Oct 2013 10:22:06 -0700
From:	"H. Peter Anvin" <hpa@...or.com>
To:	Ingo Molnar <mingo@...nel.org>
CC:	Peter Zijlstra <peterz@...radead.org>,
	Adrian Hunter <adrian.hunter@...el.com>,
	linux-kernel@...r.kernel.org, tglx@...utronix.de,
	linux-tip-commits@...r.kernel.org
Subject: Re: [tip:perf/core] perf/x86: Clean up cap_user_time* setting

On 10/06/2013 02:10 AM, Ingo Molnar wrote:
>>
>> I'm wanting to hear from the x86 people on why we have this absurd knob 
>> to begin with; but I'm tempted to simply disable all of perf if you 
>> touch it.
> 
> I'm fully with you, please zap the 'notsc' boot option - it's an ancient 
> relic, if any box is still broken with the TSC on we want to hear about it 
> and fix it!
> 

Perhaps better would be to make the notsc option do what other feature
removal options do and just remove the CPU feature flag.

Early on we had a bunch of ad hoc behaviors for feature disabling.  They
are harmful and just wrong... "not present" and "disabled" should be the
same thing in 99% of all cases (in the case of the TSC one may wish to
set the CR4 bit which disables the TSC from userspace, but I don't think
"notsc" ever did that.)

However:

pr_warn("Kernel compiled with CONFIG_X86_TSC, cannot disable TSC
completely\n");

That is a total "say what"?

At one point it even said:

printk(KERN_WARNING "notsc: Kernel compiled with CONFIG_X86_TSC, "	
"cannot disable TSC.\n");

CONFIG_X86_TSC is a baseline control option; we shouldn't key
functionality off of it.  It's fine to say notsc -> no tracing, but
making it a compile-time key makes me a bit uphappy.  We cut off 386,
but cutting of 486 at this point makes me nervous.

	-hpa

--
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/