linux-kernel - Re: [PATCH -tip] x86: cpu_debug.c prepare report if files are inappropriate or CPU is not supported

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: <20090415101700.GE6669@elte.hu>
Date:	Wed, 15 Apr 2009 12:17:00 +0200
From:	Ingo Molnar <mingo@...e.hu>
To:	Jaswinder Singh Rajput <jaswinder@...nel.org>
Cc:	Alan Cox <alan@...rguk.ukuu.org.uk>,
	x86 maintainers <x86@...nel.org>,
	LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH -tip] x86: cpu_debug.c prepare report if files are
	inappropriate or CPU is not supported


* Jaswinder Singh Rajput <jaswinder@...nel.org> wrote:

> On Tue, 2009-04-14 at 21:16 +0200, Ingo Molnar wrote:
> > * Alan Cox <alan@...rguk.ukuu.org.uk> wrote:
> > 
> > > > > Do you still think that boot_cpu_data.x86_vendor is better option in
> > > > > case for multiple CPUs.
> > > > 
> > > > yes. Assymetric SMP never really happened on x86.
> > > 
> > > It did but not between vendors. You can get away with a vendor 
> > > assumption but cpu type (mixed 486SX/DX, PII/PIII/Celeron) mixed 
> > > steppings and mixed speeds do occur. We've never supported the 
> > > mixed 486SX cases but the PII/PIII cases work (or at least 
> > > worked).
> > 
> > yeah - but look at the specific purpose here: we are deciding 
> > whether to print out state information related to major CPU 
> > features. Mixed steppings/speeds might happen, mixed apic / 
> > non-lapic not really.
> > 
> 
> We are trying to debug each CPU, so we should also collect information
> from each CPU.
> 
> Just reading information from boot CPU and display information for 
> all CPU is not correct. If we are getting information from boot 
> CPU then we only need to show information for boot CPU and ignore 
> others.
> 
> I am trying to adding more features which will be unique for each 
> CPU/core.
> 
> So it seems current model is correct where I am collecting 
> information for each CPU and printing information for each CPU.

that's OK - and we have all the per cpu data too.

The main beef i had with your code is that it copies CPU enumeration 
over into some local variables (cpu_modelflag, cpu_model) with 
redundant encodings and decodings which dont fully work.

Is there anything that your encoding/decoding does that cannot be 
done via the standard methods?

Do:

   git grep X86_VENDOR_INTEL arch/x86/

To see existing coding practices that make use of these facilities.

	Ingo
--
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/