linux-kernel - Re: [patch 21/24] perfmon: Intel architectural PMU support (x86)

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: <20081126164511.GA6703@one.firstfloor.org>
Date:	Wed, 26 Nov 2008 17:45:11 +0100
From:	Andi Kleen <andi@...stfloor.org>
To:	eranian@...il.com
Cc:	Thomas Gleixner <tglx@...utronix.de>, linux-kernel@...r.kernel.org,
	akpm@...ux-foundation.org, mingo@...e.hu, x86@...nel.org,
	andi@...stfloor.org, sfr@...b.auug.org.au
Subject: Re: [patch 21/24] perfmon: Intel architectural PMU support (x86)

> Both perfmon_intel_arch.c and perfmon_amd64.c are supposed to be kernel modules.
> They are hardcoded right now to make the patch simpler. Have PMU description be
> modules is a key features because it allows adding new processor support without
> necessarily patch the core kernel or rebooting. This has been working

To install the new processor in the first place you have 
to reboot anyways.

Also typically at least for new families (which tend to be the
only ones with radically new performance counters) it's typically
needed to change some things in the core kernel anyways. So 
this doesn't seem like a really useful feature.

> nicely on Itanium.
> With the introduction of Intel architectural Perfmon (IA32 SDM chapter
> 18), this becomes
> possible on Intel X86 as well.

It becomes possible, but without having to use any modules. It should
just work.

Probably even without it worked -- at least if you limit yourself to 
family 6 -- because the register layout all stayed the same too.

That said having modular PMUs is probably a good thing for
distribution kernels, but there is really no need for any 
code compromises just to avoid a core kernel patch now and then.

-Andi
--
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/