From: "Radim Krčmář" <rkrcmar@redhat.com>
To: Borislav Petkov <bp@alien8.de>
Cc: Nadav Amit <namit@cs.technion.ac.il>,
Ingo Molnar <mingo@kernel.org>,
Paolo Bonzini <pbonzini@redhat.com>,
"H. Peter Anvin" <hpa@zytor.com>, Ingo Molnar <mingo@redhat.com>,
Thomas Gleixner <tglx@linutronix.de>,
the arch/x86 maintainers <x86@kernel.org>,
kvm <kvm@vger.kernel.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
Andrew Morton <akpm@linux-foundation.org>,
Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: Re: [RESEND PATCH 1/3] x86: Adding structs to reflect cpuid fields
Date: Thu, 18 Sep 2014 02:29:54 +0200 [thread overview]
Message-ID: <20140918002953.GA6918@potion.redhat.com> (raw)
In-Reply-To: <20140917152221.GF5358@nazgul.tnic>
2014-09-17 17:22+0200, Borislav Petkov:
> On Wed, Sep 17, 2014 at 05:04:33PM +0200, Radim Krčmář wrote:
> > which would result in a similar if-else hack
> >
> > if (family > X)
> > ebx.split.max_monitor_line_size_after_family_X = 0
> > else
> > ebx.split.max_monitor_line_size = 0
> >
> > other options are
> > ebx.split.after_family_X.max_monitor_line_size
> > or even
> > ebx.split.max_monitor_line_size.after_family_X
>
> And how is that better than simply doing
>
> cpuid = cpuid_ebx(5);
>
> if (family > X)
> max_monitor_line_size = cpuid & MASK_FAM_X;
> else
> max_monitor_line_size = cpuid & MASK_BEFORE_FAM_X;
>
> ?
>
> With proper variable naming all is perfectly clear, readable
> and simple. You don't need to open even the CPUID manual - the
> variable tells you you're getting the max monitor line size -
> "ebx.split.max_monitor_line_size_after_family_X" needs me to parse it
> with my eyes first.
I think you proposed to use magic constant in place of of MASK_FAM_X, so
the code above is
if (family > X)
max_monitor_line_size = cpuid & 0x1ffff;
else
max_monitor_line_size = cpuid & 0xffff;
We can nicely oneline it, but that's about all the benefits I can see.
It is prone to typos, hard to search for and limiting our operations to
a simple assignment to a properly named variable.
(I prefer descriptive, horribly long, names to raw constant everywhere,
MASK_MAX_MONITOR_LINE_SIZE_FAM_X.)
Second problem: Most elements don't begin at offset 0, so the usual
retrieval would add a shift, (repurposing max_monitor_line_size)
max_monitor_line_size = (cpuid & MASK_FAM_X) >> OFFSET_FAM_X;
and it's not better when we write it back after doing stuff.
cpuid = (cpuid & ~MASK_FAM_X) | (max_monitor_line_size << OFFSET_FAM_X
& MASK_FAM_X);
All would be fine if we abstracted this with more macros ... wait,
bitfield already does that!
max_monitor_line_size = cpuid.split.max_monitor_line_size_fam_x;
cpuid.split.max_monitor_line_size_fam_x = max_monitor_line_size;
---
OT: I'd remove '.split', but we probably wouldn't agree about '.full'.
next prev parent reply other threads:[~2014-09-18 0:30 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1410870160-28845-1-git-send-email-namit@cs.technion.ac.il>
2014-09-16 13:22 ` [PATCH 0/3] x86: structs for cpuid info in x86 Ingo Molnar
2014-09-16 20:19 ` Nadav Amit
2014-09-17 12:37 ` Ingo Molnar
2014-09-17 12:45 ` Borislav Petkov
2014-09-17 12:54 ` [RESEND PATCH " Nadav Amit
2014-09-17 12:54 ` [RESEND PATCH 1/3] x86: Adding structs to reflect cpuid fields Nadav Amit
2014-09-17 13:21 ` Borislav Petkov
2014-09-17 13:53 ` Nadav Amit
2014-09-17 14:06 ` Borislav Petkov
2014-09-17 15:04 ` Radim Krčmář
2014-09-17 15:22 ` Borislav Petkov
2014-09-18 0:29 ` Radim Krčmář [this message]
2014-09-18 7:19 ` Borislav Petkov
2014-09-18 10:00 ` Radim Krčmář
2014-09-18 13:06 ` Paolo Bonzini
2014-09-18 13:26 ` Borislav Petkov
2014-09-18 13:36 ` Paolo Bonzini
2014-09-19 7:58 ` Borislav Petkov
2014-09-19 8:59 ` Nadav Amit
2014-09-19 10:32 ` Borislav Petkov
2014-09-19 13:40 ` Paolo Bonzini
2014-09-19 14:44 ` Borislav Petkov
2014-09-17 14:10 ` Peter Zijlstra
2014-09-17 12:54 ` [RESEND PATCH 2/3] x86: Use new cpuid structs in cpuid functions Nadav Amit
2014-09-17 12:54 ` [RESEND PATCH 3/3] KVM: x86: Using cpuid structs in KVM Nadav Amit
2014-09-17 14:12 ` [PATCH 0/3] x86: structs for cpuid info in x86 Peter Zijlstra
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20140918002953.GA6918@potion.redhat.com \
--to=rkrcmar@redhat.com \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=bp@alien8.de \
--cc=hpa@zytor.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=mingo@redhat.com \
--cc=namit@cs.technion.ac.il \
--cc=pbonzini@redhat.com \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).