linux-kernel - Re: [PATCH 2/3] x86: Implement prctl PR_GET_TSC and PR_SET

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]

Date:	Sat, 12 Apr 2008 14:12:46 -0700
From:	Arjan van de Ven <arjan@...radead.org>
To:	Erik Bosman <ejbosman@...vu.nl>
Cc:	"H. Peter Anvin" <hpa@...or.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...hat.com>,
	Andrea Arcangeli <andrea@...share.com>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/3] x86: Implement prctl PR_GET_TSC and PR_SET_TSC

On Sat, 12 Apr 2008 23:05:44 +0200 (CEST)
Erik Bosman <ejbosman@...vu.nl> wrote:

> 
> 
> On Sat, 12 Apr 2008, H. Peter Anvin wrote:
> 
> > Arjan van de Ven wrote:
> > >
> > > Hi,
> > >
> > > why did you make this a configuration option? In general it's not
> > > a good idea to make userspace visible ABI's (PR_* clearly is one
> > > of these) a CONFIG option unless there's some HUGE space saving
> > > going on. I don't see that here....
> > >
> > > So can you explain your rationale for making this a config option?
> > >
> 
> The ABI itself is not a config option (see patch 1/3.)
> If the x86 implementation patch isn't applied, prctl() will
> return -EINVAL, just like most other PR_* options, which are
> only implemented on specific architectures, take
> (PR_GET/PR_SET_ENDIAN) as an example.

that doesn't say why you made it an option still :)
Something like this (which looks like generally useful functionality)
should just be there unless there is a really high cost....

> 
> >
> > I also saw no mention about performance impact, which need to be
> > considered whenever *anything* is proposed to be inserted into a hot
> > path.  It may be (heck, *should be*) that the performance impact
> > isn't measurable, but that needs to be positively established.
> >
> 
> This is why I made the implementation part configurable, although I
> don't think the overhead will be very high since it seems to me that
> the __switch_to_xtra function was designed explicitly to keep unusual
> options out of the hot path.

so you normally don't have performance overhead unless you use the feature....
that's really good!
It also means that there wouldn't be a need for this to be a config option, but
instead it can just always be there.. saving us a ton of ifdefs as well as the burden
of having an extra config option.

-- 
If you want to reach me at my work email, use arjan@...ux.intel.com
For development, discussion and tips for power savings, 
visit http://www.lesswatts.org
--
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/